diff options
Diffstat (limited to 'tests')
24 files changed, 239 insertions, 1121 deletions
diff --git a/tests/vpp/perf/__init__.robot b/tests/vpp/perf/__init__.robot index 25ed403d8a..1f65414306 100644 --- a/tests/vpp/perf/__init__.robot +++ b/tests/vpp/perf/__init__.robot @@ -37,14 +37,10 @@ | | ... | Setup suite Variables. Variables are used across performance testing. | | ... | | ... | _NOTE:_ This KW sets following suite variables: -| | ... | - perf_vm_image - Guest VM disk image -| | ... | - perf_qemu_path - Path prefix to QEMU binary -| | ... | - use_tuned_cfs - Switch to set scheduler policy -| | ... | - qemu_build - Whether Qemu will be built -| | ... | - pkt_trace - Switch to enable packet trace for test -| | ... | - dut_stats - Switch to enable DUT statistics -| | ... | - uio_driver - Default UIO driver -| | ... | - plugins_to_enable - List of plugins to be enabled for test +| | ... | - pkt_trace - Switch to enable packet trace for test. +| | ... | - dut_stats - Switch to enable DUT statistics. +| | ... | - plugins_to_enable - List of plugins to be enabled for test. +| | ... | - packages_dir - Directory with VPP binary packages. | | ... | | Set Global Variable | ${pkt_trace} | ${False} | | Set Global Variable | ${dut_stats} | ${True} diff --git a/tests/vpp/perf/vm_vhost/10ge2p1x710-1lbvpplacp-dot1q-l2bdbasemaclrn-eth-2vhostvr1024-1vm-ndrpdr.robot b/tests/vpp/perf/vm_vhost/10ge2p1x710-1lbvpplacp-dot1q-l2bdbasemaclrn-eth-2vhostvr1024-1vm-ndrpdr.robot index 504320a8b2..6040a320f2 100644 --- a/tests/vpp/perf/vm_vhost/10ge2p1x710-1lbvpplacp-dot1q-l2bdbasemaclrn-eth-2vhostvr1024-1vm-ndrpdr.robot +++ b/tests/vpp/perf/vm_vhost/10ge2p1x710-1lbvpplacp-dot1q-l2bdbasemaclrn-eth-2vhostvr1024-1vm-ndrpdr.robot @@ -26,13 +26,12 @@ | Suite Teardown | Tear down 3-node performance topology | ... | Test Setup | Set up performance test -| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd -| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs} -| ... | dut2_node=${dut2} | dut2_vm_refs=${dut2_vm_refs} +| Test Teardown | Tear down performance test with vhost | ... | Test Template | Local Template | ... -| Documentation | *RFC2544: Pkt throughput L2BD test cases with vhost and vpp link bonding* +| Documentation | *RFC2544: Pkt throughput L2BD test cases with vhost and vpp +| ... | link bonding* | ... | ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology | ... | with single links between nodes. @@ -40,13 +39,10 @@ | ... | tagging is applied on link between DUT1 and DUT2. | ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with VPP | ... | link bonding (mode LACP, transmit policy l34) on link between DUT1 and -| ... | DUT2 and L2 bridge-domain with MAC learning enabled. Qemu Guest is -| ... | connected to VPP via vhost-user interfaces. Guest is running DPDK -| ... | testpmd interconnecting vhost-user interfaces using 5 cores pinned to -| ... | cpus 5-9 and 2048M memory. Testpmd is using socket-mem=1024M (512x2M -| ... | hugepages), 5 cores (1 main core and 4 cores dedicated for io), -| ... | forwarding mode is set to io, rxd/txd=1024, burst=64. DUT1, DUT2 are -| ... | tested with ${nic_name}.\ +| ... | DUT2 and L2 bridge-domain with MAC learning enabled. Qemu VNFs are \ +| ... | connected to VPP via vhost-user interfaces. Guest is running VPP l2xc \ +| ... | interconnecting vhost-user interfaces, rxd/txd=1024. DUT1/DUT2 is \ +| ... | tested with ${nic_name}. | ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop\ | ... | Rate) with zero packet loss tolerance and throughput PDR (Partial Drop\ | ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage\ @@ -57,12 +53,17 @@ | ... | flow-group) with all packets containing Ethernet header, IPv4 header | ... | with IP protocol=61 and static payload. MAC addresses are matching MAC | ... | addresses of the TG node interfaces. +| ... | *[Ref] Applicable standard specifications:* RFC2544. *** Variables *** | ${nic_name}= | Intel-X710 | ${overhead}= | ${4} | ${subid}= | 10 | ${tag_rewrite}= | pop-1 +| ${nf_dtcr}= | ${1} +| ${nf_dtc}= | ${1} +| ${nf_chains}= | ${1} +| ${nf_nodes}= | ${1} # Link bonding config | ${bond_mode}= | lacp | ${lb_mode}= | l34 @@ -88,10 +89,6 @@ | | ... | | Set Test Variable | \${frame_size} | | ... -| | ${dut1_vm_refs}= | Create Dictionary -| | ${dut2_vm_refs}= | Create Dictionary -| | Set Test Variable | ${dut1_vm_refs} -| | Set Test Variable | ${dut2_vm_refs} | | Given Add worker threads and rxqueues to all DUTs | ${phy_cores} | ${rxq} | | And Add PCI devices to all DUTs | | And Add VLAN Strip Offload switch off between DUTs in 3-node single link topology @@ -100,10 +97,9 @@ | | When Initialize L2 bridge domains with Vhost-User and VLAN with VPP link bonding in a 3-node circular topology | | ... | ${bd_id1} | ${bd_id2} | ${subid} | | ... | ${tag_rewrite} | ${bond_mode} | ${lb_mode} -| | And Configure guest VMs with dpdk-testpmd connected via vhost-user -| | ... | vm_count=${1} | jumbo=${jumbo} | perf_qemu_qsz=${1024} -| | ... | use_tuned_cfs=${False} -| | And All Vpp Interfaces Ready Wait | ${nodes} +| | And Configure chains of NFs connected via vhost-user +| | ... | nf_chains=${nf_chains} | nf_nodes=${nf_nodes} | jumbo=${jumbo} +| | ... | use_tuned_cfs=${False} | auto_scale=${True} | vnf=vpp_chain_l2xc | | Then Find NDR and PDR intervals using optimized search *** Test Cases *** diff --git a/tests/vpp/perf/vm_vhost/10ge2p1x710-1lbvpplacp-dot1q-l2xcbase-eth-2vhostvr1024-1vm-ndrpdr.robot b/tests/vpp/perf/vm_vhost/10ge2p1x710-1lbvpplacp-dot1q-l2xcbase-eth-2vhostvr1024-1vm-ndrpdr.robot index 7051a86140..8807dc5184 100644 --- a/tests/vpp/perf/vm_vhost/10ge2p1x710-1lbvpplacp-dot1q-l2xcbase-eth-2vhostvr1024-1vm-ndrpdr.robot +++ b/tests/vpp/perf/vm_vhost/10ge2p1x710-1lbvpplacp-dot1q-l2xcbase-eth-2vhostvr1024-1vm-ndrpdr.robot @@ -26,13 +26,12 @@ | Suite Teardown | Tear down 3-node performance topology | ... | Test Setup | Set up performance test -| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd -| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs} -| ... | dut2_node=${dut2} | dut2_vm_refs=${dut2_vm_refs} +| Test Teardown | Tear down performance test with vhost | ... | Test Template | Local Template | ... -| Documentation | *RFC2544: Pkt throughput L2XC test cases with vhost and vpp link bonding* +| Documentation | *RFC2544: Pkt throughput L2XC test cases with vhost and vpp +| ... | link bonding* | ... | ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology | ... | with single links between nodes. @@ -40,12 +39,10 @@ | ... | tagging is applied on link between DUT1 and DUT2. | ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with VPP | ... | link bonding (mode LACP, transmit policy l34) on link between DUT1 and -| ... | DUT2 and L2 cross- connect. Qemu Guest is connected to VPP via -| ... | vhost-user interfaces. Guest is running DPDK testpmd interconnecting -| ... | vhost-user interfaces using 5 cores pinned to cpus 5-9 and 2048M memory. -| ... | Testpmd is using socket-mem=1024M (512x2M hugepages), 5 cores (1 main -| ... | core and 4 cores dedicated for io), forwarding mode is set to io, -| ... | rxd/txd=1024, burst=64. DUT1, DUT2 are tested with ${nic_name}.\ +| ... | DUT2 and L2 cross- connect. Qemu VNFs are \ +| ... | connected to VPP via vhost-user interfaces. Guest is running VPP l2xc \ +| ... | interconnecting vhost-user interfaces, rxd/txd=1024. DUT1/DUT2 is \ +| ... | tested with ${nic_name}. | ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop\ | ... | Rate) with zero packet loss tolerance and throughput PDR (Partial Drop\ | ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage\ @@ -56,12 +53,17 @@ | ... | flow-group) with all packets containing Ethernet header, IPv4 header | ... | with IP protocol=61 and static payload. MAC addresses are matching MAC | ... | addresses of the TG node interfaces. +| ... | *[Ref] Applicable standard specifications:* RFC2544. *** Variables *** | ${nic_name}= | Intel-X710 | ${overhead}= | ${4} | ${subid}= | 10 | ${tag_rewrite}= | pop-1 +| ${nf_dtcr}= | ${1} +| ${nf_dtc}= | ${1} +| ${nf_chains}= | ${1} +| ${nf_nodes}= | ${1} # Link bonding config | ${bond_mode}= | lacp | ${lb_mode}= | l34 @@ -85,10 +87,6 @@ | | ... | | Set Test Variable | \${frame_size} | | ... -| | ${dut1_vm_refs}= | Create Dictionary -| | ${dut2_vm_refs}= | Create Dictionary -| | Set Test Variable | ${dut1_vm_refs} -| | Set Test Variable | ${dut2_vm_refs} | | Given Add worker threads and rxqueues to all DUTs | ${phy_cores} | ${rxq} | | And Add PCI devices to all DUTs | | And Add VLAN Strip Offload switch off between DUTs in 3-node single link topology @@ -96,10 +94,9 @@ | | And Apply startup configuration on all VPP DUTs | | When Initialize L2 xconnect with Vhost-User and VLAN with VPP link bonding in 3-node circular topology | | ... | ${subid} | ${tag_rewrite} | ${bond_mode} | ${lb_mode} -| | And Configure guest VMs with dpdk-testpmd connected via vhost-user -| | ... | vm_count=${1} | jumbo=${jumbo} | perf_qemu_qsz=${1024} -| | ... | use_tuned_cfs=${False} -| | And All Vpp Interfaces Ready Wait | ${nodes} +| | And Configure chains of NFs connected via vhost-user +| | ... | nf_chains=${nf_chains} | nf_nodes=${nf_nodes} | jumbo=${jumbo} +| | ... | use_tuned_cfs=${False} | auto_scale=${True} | vnf=vpp_chain_l2xc | | Then Find NDR and PDR intervals using optimized search *** Test Cases *** diff --git a/tests/vpp/perf/vm_vhost/10ge2p1x710-2lbvpplacp-dot1q-l2bdbasemaclrn-eth-2vhostvr1024-1vm-ndrpdr.robot b/tests/vpp/perf/vm_vhost/10ge2p1x710-2lbvpplacp-dot1q-l2bdbasemaclrn-eth-2vhostvr1024-1vm-ndrpdr.robot index f3e4a114f4..05925de881 100644 --- a/tests/vpp/perf/vm_vhost/10ge2p1x710-2lbvpplacp-dot1q-l2bdbasemaclrn-eth-2vhostvr1024-1vm-ndrpdr.robot +++ b/tests/vpp/perf/vm_vhost/10ge2p1x710-2lbvpplacp-dot1q-l2bdbasemaclrn-eth-2vhostvr1024-1vm-ndrpdr.robot @@ -26,13 +26,12 @@ | Suite Teardown | Tear down 3-node performance topology | ... | Test Setup | Set up performance test -| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd -| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs} -| ... | dut2_node=${dut2} | dut2_vm_refs=${dut2_vm_refs} +| Test Teardown | Tear down performance test with vhost | ... | Test Template | Local Template | ... -| Documentation | *RFC2544: Pkt throughput L2BD test cases with vhost and vpp link bonding* +| Documentation | *RFC2544: Pkt throughput L2BD test cases with vhost and vpp +| ... | link bonding* | ... | ... | *[Top] Network Topologies:* TG-DUT1=DUT2-TG 3-node circular topology | ... | with single links between TG and DUT nodes and double link between DUT @@ -41,13 +40,10 @@ | ... | tagging is applied on link between DUT1 and DUT2. | ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with VPP | ... | link bonding (mode LACP, transmit policy l34) on link between DUT1 and -| ... | DUT2 and L2 bridge-domain with MAC learning enabled. Qemu Guest is -| ... | connected to VPP via vhost-user interfaces. Guest is running DPDK -| ... | testpmd interconnecting vhost-user interfaces using 5 cores pinned to -| ... | cpus 5-9 and 2048M memory. Testpmd is using socket-mem=1024M (512x2M -| ... | hugepages), 5 cores (1 main core and 4 cores dedicated for io), -| ... | forwarding mode is set to io, rxd/txd=1024, burst=64. DUT1, DUT2 are -| ... | tested with ${nic_name}.\ +| ... | DUT2 and L2 bridge-domain with MAC learning enabled. Qemu VNFs are \ +| ... | connected to VPP via vhost-user interfaces. Guest is running VPP l2xc \ +| ... | interconnecting vhost-user interfaces, rxd/txd=1024. DUT1/DUT2 is \ +| ... | tested with ${nic_name}. | ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop\ | ... | Rate) with zero packet loss tolerance and throughput PDR (Partial Drop\ | ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage\ @@ -58,12 +54,17 @@ | ... | flow-group) with all packets containing Ethernet header, IPv4 header | ... | with IP protocol=61 and static payload. MAC addresses are matching MAC | ... | addresses of the TG node interfaces. +| ... | *[Ref] Applicable standard specifications:* RFC2544. *** Variables *** | ${nic_name}= | Intel-X710 | ${overhead}= | ${4} | ${subid}= | 10 | ${tag_rewrite}= | pop-1 +| ${nf_dtcr}= | ${1} +| ${nf_dtc}= | ${1} +| ${nf_chains}= | ${1} +| ${nf_nodes}= | ${1} # Link bonding config | ${bond_mode}= | lacp | ${lb_mode}= | l34 @@ -89,10 +90,6 @@ | | ... | | Set Test Variable | \${frame_size} | | ... -| | ${dut1_vm_refs}= | Create Dictionary -| | ${dut2_vm_refs}= | Create Dictionary -| | Set Test Variable | ${dut1_vm_refs} -| | Set Test Variable | ${dut2_vm_refs} | | Given Add worker threads and rxqueues to all DUTs | ${phy_cores} | ${rxq} | | And Add PCI devices to all DUTs | | And Add VLAN Strip Offload switch off between DUTs in 3-node double link topology @@ -101,10 +98,9 @@ | | When Initialize L2 bridge domains with Vhost-User and VLAN with VPP link bonding in a 3-node circular topology | | ... | ${bd_id1} | ${bd_id2} | ${subid} | | ... | ${tag_rewrite} | ${bond_mode} | ${lb_mode} -| | And Configure guest VMs with dpdk-testpmd connected via vhost-user -| | ... | vm_count=${1} | jumbo=${jumbo} | perf_qemu_qsz=${1024} -| | ... | use_tuned_cfs=${False} -| | And All Vpp Interfaces Ready Wait | ${nodes} +| | And Configure chains of NFs connected via vhost-user +| | ... | nf_chains=${nf_chains} | nf_nodes=${nf_nodes} | jumbo=${jumbo} +| | ... | use_tuned_cfs=${False} | auto_scale=${True} | vnf=vpp_chain_l2xc | | Then Find NDR and PDR intervals using optimized search *** Test Cases *** diff --git a/tests/vpp/perf/vm_vhost/10ge2p1x710-2lbvpplacp-dot1q-l2xcbase-eth-2vhostvr1024-1vm-ndrpdr.robot b/tests/vpp/perf/vm_vhost/10ge2p1x710-2lbvpplacp-dot1q-l2xcbase-eth-2vhostvr1024-1vm-ndrpdr.robot index 5efe46bbb2..5265c18f58 100644 --- a/tests/vpp/perf/vm_vhost/10ge2p1x710-2lbvpplacp-dot1q-l2xcbase-eth-2vhostvr1024-1vm-ndrpdr.robot +++ b/tests/vpp/perf/vm_vhost/10ge2p1x710-2lbvpplacp-dot1q-l2xcbase-eth-2vhostvr1024-1vm-ndrpdr.robot @@ -26,13 +26,12 @@ | Suite Teardown | Tear down 3-node performance topology | ... | Test Setup | Set up performance test -| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd -| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs} -| ... | dut2_node=${dut2} | dut2_vm_refs=${dut2_vm_refs} +| Test Teardown | Tear down performance test with vhost | ... | Test Template | Local Template | ... -| Documentation | *RFC2544: Pkt throughput L2XC test cases with vhost and vpp link bonding* +| Documentation | *RFC2544: Pkt throughput L2XC test cases with vhost and vpp +| ... | link bonding* | ... | ... | *[Top] Network Topologies:* TG-DUT1=DUT2-TG 3-node circular topology | ... | with single links between TG and DUT nodes and double link between DUT @@ -41,12 +40,10 @@ | ... | tagging is applied on link between DUT1 and DUT2. | ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with VPP | ... | link bonding (mode LACP, transmit policy l34) on link between DUT1 and -| ... | DUT2 and L2 cross- connect. Qemu Guest is connected to VPP via -| ... | vhost-user interfaces. Guest is running DPDK testpmd interconnecting -| ... | vhost-user interfaces using 5 cores pinned to cpus 5-9 and 2048M memory. -| ... | Testpmd is using socket-mem=1024M (512x2M hugepages), 5 cores (1 main -| ... | core and 4 cores dedicated for io), forwarding mode is set to io, -| ... | rxd/txd=1024, burst=64. DUT1, DUT2 are tested with ${nic_name}.\ +| ... | DUT2 and L2 cross- connect. Qemu VNFs are \ +| ... | connected to VPP via vhost-user interfaces. Guest is running VPP l2xc \ +| ... | interconnecting vhost-user interfaces, rxd/txd=1024. DUT1/DUT2 is \ +| ... | tested with ${nic_name}. | ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop\ | ... | Rate) with zero packet loss tolerance and throughput PDR (Partial Drop\ | ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage\ @@ -57,12 +54,17 @@ | ... | flow-group) with all packets containing Ethernet header, IPv4 header | ... | with IP protocol=61 and static payload. MAC addresses are matching MAC | ... | addresses of the TG node interfaces. +| ... | *[Ref] Applicable standard specifications:* RFC2544. *** Variables *** | ${nic_name}= | Intel-X710 | ${overhead}= | ${4} | ${subid}= | 10 | ${tag_rewrite}= | pop-1 +| ${nf_dtcr}= | ${1} +| ${nf_dtc}= | ${1} +| ${nf_chains}= | ${1} +| ${nf_nodes}= | ${1} # Link bonding config | ${bond_mode}= | lacp | ${lb_mode}= | l34 @@ -86,10 +88,6 @@ | | ... | | Set Test Variable | \${frame_size} | | ... -| | ${dut1_vm_refs}= | Create Dictionary -| | ${dut2_vm_refs}= | Create Dictionary -| | Set Test Variable | ${dut1_vm_refs} -| | Set Test Variable | ${dut2_vm_refs} | | Given Add worker threads and rxqueues to all DUTs | ${phy_cores} | ${rxq} | | And Add PCI devices to all DUTs | | And Add VLAN Strip Offload switch off between DUTs in 3-node double link topology @@ -97,10 +95,9 @@ | | And Apply startup configuration on all VPP DUTs | | When Initialize L2 xconnect with Vhost-User and VLAN with VPP link bonding in 3-node circular topology | | ... | ${subid} | ${tag_rewrite} | ${bond_mode} | ${lb_mode} -| | And Configure guest VMs with dpdk-testpmd connected via vhost-user -| | ... | vm_count=${1} | jumbo=${jumbo} | perf_qemu_qsz=${1024} -| | ... | use_tuned_cfs=${False} -| | And All Vpp Interfaces Ready Wait | ${nodes} +| | And Configure chains of NFs connected via vhost-user +| | ... | nf_chains=${nf_chains} | nf_nodes=${nf_nodes} | jumbo=${jumbo} +| | ... | use_tuned_cfs=${False} | auto_scale=${True} | vnf=vpp_chain_l2xc | | Then Find NDR and PDR intervals using optimized search *** Test Cases *** diff --git a/tests/vpp/perf/vm_vhost/10ge2p1x710-dot1q-l2bdbasemaclrn-eth-2vhostvr1024-1vm-ndrpdr.robot b/tests/vpp/perf/vm_vhost/10ge2p1x710-dot1q-l2bdbasemaclrn-eth-2vhostvr1024-1vm-ndrpdr.robot index 304b03d6bb..ea2a5834c2 100644 --- a/tests/vpp/perf/vm_vhost/10ge2p1x710-dot1q-l2bdbasemaclrn-eth-2vhostvr1024-1vm-ndrpdr.robot +++ b/tests/vpp/perf/vm_vhost/10ge2p1x710-dot1q-l2bdbasemaclrn-eth-2vhostvr1024-1vm-ndrpdr.robot @@ -23,9 +23,7 @@ | Suite Teardown | Tear down 3-node performance topology | ... | Test Setup | Set up performance test -| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd -| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs} -| ... | dut2_node=${dut2} | dut2_vm_refs=${dut2_vm_refs} +| Test Teardown | Tear down performance test with vhost | ... | Test Template | Local Template | ... @@ -37,10 +35,10 @@ | ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 switching of IPv4. IEEE\ | ... | 802.1Q tagging is applied on link between DUT1 and DUT2. | ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 bridge-\ -| ... | domain and MAC learning enabled. Qemu Guest is connected to VPP via\ -| ... | vhost-user interfaces. Guest is running DPDK testpmd interconnecting\ -| ... | vhost-user interfaces, forwarding mode is set to io, rxd/txd=1024.\ -| ... | DUT1 and DUT2 are tested with ${nic_name}.\ +| ... | domain and MAC learning enabled. Qemu VNFs are \ +| ... | connected to VPP via vhost-user interfaces. Guest is running VPP l2xc \ +| ... | interconnecting vhost-user interfaces, rxd/txd=1024. DUT1/DUT2 is \ +| ... | tested with ${nic_name}. | ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop\ | ... | Rate) with zero packet loss tolerance and throughput PDR (Partial Drop\ | ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage\ @@ -61,6 +59,10 @@ | ${tag_rewrite}= | pop-1 | ${bd_id1}= | 1 | ${bd_id2}= | 2 +| ${nf_dtcr}= | ${1} +| ${nf_dtc}= | ${1} +| ${nf_chains}= | ${1} +| ${nf_nodes}= | ${1} # Traffic profile: | ${traffic_profile}= | trex-sl-3n-ethip4-ip4src254 @@ -81,19 +83,15 @@ | | ... | | Set Test Variable | \${frame_size} | | ... -| | ${dut1_vm_refs}= | Create Dictionary -| | ${dut2_vm_refs}= | Create Dictionary -| | Set Test Variable | ${dut1_vm_refs} -| | Set Test Variable | ${dut2_vm_refs} | | Given Add worker threads and rxqueues to all DUTs | ${phy_cores} | ${rxq} | | And Add PCI devices to all DUTs | | Set Max Rate And Jumbo And Handle Multi Seg | | And Apply startup configuration on all VPP DUTs | | When Initialize L2 bridge domains with Vhost-User and VLAN in circular topology | | ... | ${bd_id1} | ${bd_id2} | ${subid} | ${tag_rewrite} -| | And Configure guest VMs with dpdk-testpmd connected via vhost-user -| | ... | vm_count=${1} | jumbo=${jumbo} | perf_qemu_qsz=${1024} -| | ... | use_tuned_cfs=${False} +| | And Configure chains of NFs connected via vhost-user +| | ... | nf_chains=${nf_chains} | nf_nodes=${nf_nodes} | jumbo=${jumbo} +| | ... | use_tuned_cfs=${False} | auto_scale=${True} | vnf=vpp_chain_l2xc | | Then Find NDR and PDR intervals using optimized search *** Test Cases *** diff --git a/tests/vpp/perf/vm_vhost/10ge2p1x710-dot1q-l2xcbase-eth-2vhostvr1024-1vm-ndrpdr.robot b/tests/vpp/perf/vm_vhost/10ge2p1x710-dot1q-l2xcbase-eth-2vhostvr1024-1vm-ndrpdr.robot index 1f1d6e3890..76c613a6cb 100644 --- a/tests/vpp/perf/vm_vhost/10ge2p1x710-dot1q-l2xcbase-eth-2vhostvr1024-1vm-ndrpdr.robot +++ b/tests/vpp/perf/vm_vhost/10ge2p1x710-dot1q-l2xcbase-eth-2vhostvr1024-1vm-ndrpdr.robot @@ -22,9 +22,7 @@ | Suite Teardown | Tear down 3-node performance topology | ... | Test Setup | Set up performance test -| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd -| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs} -| ... | dut2_node=${dut2} | dut2_vm_refs=${dut2_vm_refs} +| Test Teardown | Tear down performance test with vhost | ... | Test Template | Local Template | ... @@ -35,12 +33,10 @@ | ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 cross connect. 802.1q | ... | tagging is applied on link between DUT1 and DUT2. | ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 cross- -| ... | connect. Qemu Guest is connected to VPP via -| ... | vhost-user interfaces. Guest is running DPDK testpmd interconnecting -| ... | vhost-user interfaces using 5 cores pinned to cpus 5-9 and 2048M -| ... | memory. Testpmd is using socket-mem=1024M (512x2M hugepages), 5 cores -| ... | (1 main core and 4 cores dedicated for io), forwarding mode is set to -| ... | io, rxd/txd=1024, burst=64. DUT1, DUT2 are tested with ${nic_name}.\ +| ... | connect. Qemu VNFs are \ +| ... | connected to VPP via vhost-user interfaces. Guest is running VPP l2xc \ +| ... | interconnecting vhost-user interfaces, rxd/txd=1024. DUT1/DUT2 is \ +| ... | tested with ${nic_name}. | ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop\ | ... | Rate) with zero packet loss tolerance and throughput PDR (Partial Drop\ | ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage\ @@ -51,12 +47,17 @@ | ... | flow-group) with all packets containing Ethernet header, IPv4 header | ... | with IP protocol=61 and static payload. MAC addresses are matching MAC | ... | addresses of the TG node interfaces. +| ... | *[Ref] Applicable standard specifications:* RFC2544. *** Variables *** | ${nic_name}= | Intel-X710 | ${overhead}= | ${4} | ${subid}= | 10 | ${tag_rewrite}= | pop-1 +| ${nf_dtcr}= | ${1} +| ${nf_dtc}= | ${1} +| ${nf_chains}= | ${1} +| ${nf_nodes}= | ${1} # Traffic profile: | ${traffic_profile}= | trex-sl-3n-ethip4-ip4src254 @@ -77,19 +78,15 @@ | | ... | | Set Test Variable | \${frame_size} | | ... -| | ${dut1_vm_refs}= | Create Dictionary -| | ${dut2_vm_refs}= | Create Dictionary -| | Set Test Variable | ${dut1_vm_refs} -| | Set Test Variable | ${dut2_vm_refs} | | Given Add worker threads and rxqueues to all DUTs | ${phy_cores} | ${rxq} | | And Add PCI devices to all DUTs | | Set Max Rate And Jumbo And Handle Multi Seg | | And Apply startup configuration on all VPP DUTs | | When Initialize L2 xconnect with Vhost-User and VLAN in 3-node circular topology | | ... | ${subid} | ${tag_rewrite} -| | And Configure guest VMs with dpdk-testpmd connected via vhost-user -| | ... | vm_count=${1} | jumbo=${jumbo} | perf_qemu_qsz=${1024} -| | ... | use_tuned_cfs=${False} +| | And Configure chains of NFs connected via vhost-user +| | ... | nf_chains=${nf_chains} | nf_nodes=${nf_nodes} | jumbo=${jumbo} +| | ... | use_tuned_cfs=${False} | auto_scale=${True} | vnf=vpp_chain_l2xc | | Then Find NDR and PDR intervals using optimized search *** Test Cases *** diff --git a/tests/vpp/perf/vm_vhost/10ge2p1x710-eth-l2bdbasemaclrn-eth-2vhostvr1024-1vm-ndrpdr.robot b/tests/vpp/perf/vm_vhost/10ge2p1x710-eth-l2bdbasemaclrn-eth-2vhostvr1024-1vm-ndrpdr.robot index 8e9da645b0..ae426fe86d 100644 --- a/tests/vpp/perf/vm_vhost/10ge2p1x710-eth-l2bdbasemaclrn-eth-2vhostvr1024-1vm-ndrpdr.robot +++ b/tests/vpp/perf/vm_vhost/10ge2p1x710-eth-l2bdbasemaclrn-eth-2vhostvr1024-1vm-ndrpdr.robot @@ -22,9 +22,7 @@ | Suite Teardown | Tear down 3-node performance topology | ... | Test Setup | Set up performance test -| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd -| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs} -| ... | dut2_node=${dut2} | dut2_vm_refs=${dut2_vm_refs} +| Test Teardown | Tear down performance test with vhost | ... | Test Template | Local Template | ... @@ -34,12 +32,10 @@ | ... | with single links between nodes. | ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 switching of IPv4. | ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 bridge- -| ... | domain and MAC learning enabled. Qemu Guest is connected to VPP via -| ... | vhost-user interfaces. Guest is running DPDK testpmd interconnecting -| ... | vhost-user interfaces using 5 cores pinned to cpus on NUMA1 and 2048M -| ... | memory. Testpmd is using socket-mem=1024M (512x2M hugepages), 5 cores -| ... | (1 main core and 4 cores dedicated for io), forwarding mode is set to -| ... | io, rxd/txd=1024, burst=64. DUT1, DUT2 are tested with ${nic_name}.\ +| ... | domain and MAC learning enabled. Qemu VNFs are \ +| ... | connected to VPP via vhost-user interfaces. Guest is running VPP l2xc \ +| ... | interconnecting vhost-user interfaces, rxd/txd=1024. DUT1/DUT2 is \ +| ... | tested with ${nic_name}. | ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop\ | ... | Rate) with zero packet loss tolerance and throughput PDR (Partial Drop\ | ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage\ @@ -56,6 +52,10 @@ *** Variables *** | ${nic_name}= | Intel-X710 | ${overhead}= | ${0} +| ${nf_dtcr}= | ${1} +| ${nf_dtc}= | ${1} +| ${nf_chains}= | ${1} +| ${nf_nodes}= | ${1} # Traffic profile: | ${traffic_profile}= | trex-sl-3n-ethip4-ip4src254 @@ -76,18 +76,14 @@ | | ... | | Set Test Variable | \${frame_size} | | ... -| | ${dut1_vm_refs}= | Create Dictionary -| | ${dut2_vm_refs}= | Create Dictionary -| | Set Test Variable | ${dut1_vm_refs} -| | Set Test Variable | ${dut2_vm_refs} | | Given Add worker threads and rxqueues to all DUTs | ${phy_cores} | ${rxq} | | And Add PCI devices to all DUTs | | Set Max Rate And Jumbo And Handle Multi Seg | | And Apply startup configuration on all VPP DUTs -| | When Initialize L2 bridge domains with Vhost-User | nf_nodes=${1} -| | And Configure guest VMs with dpdk-testpmd connected via vhost-user -| | ... | vm_count=${1} | jumbo=${jumbo} | perf_qemu_qsz=${1024} -| | ... | use_tuned_cfs=${False} +| | When Initialize L2 bridge domains with Vhost-User | nf_nodes=${nf_nodes} +| | And Configure chains of NFs connected via vhost-user +| | ... | nf_chains=${nf_chains} | nf_nodes=${nf_nodes} | jumbo=${jumbo} +| | ... | use_tuned_cfs=${False} | auto_scale=${True} | vnf=vpp_chain_l2xc | | Then Find NDR and PDR intervals using optimized search *** Test Cases *** diff --git a/tests/vpp/perf/vm_vhost/10ge2p1x710-eth-l2bdbasemaclrn-eth-4vhostvr1024-2vm-ndrpdr.robot b/tests/vpp/perf/vm_vhost/10ge2p1x710-eth-l2bdbasemaclrn-eth-4vhostvr1024-2vm-ndrpdr.robot deleted file mode 100644 index f7222c243c..0000000000 --- a/tests/vpp/perf/vm_vhost/10ge2p1x710-eth-l2bdbasemaclrn-eth-4vhostvr1024-2vm-ndrpdr.robot +++ /dev/null @@ -1,139 +0,0 @@ -# Copyright (c) 2019 Cisco and/or its affiliates. -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at: -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. - -*** Settings *** -| Resource | resources/libraries/robot/performance/performance_setup.robot -| ... -| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDR -| ... | NIC_Intel-X710 | ETH | L2BDMACLRN | BASE | VHOST | 2VM | VHOST_1024 -| ... -| Suite Setup | Set up 3-node performance topology with DUT's NIC model -| ... | L2 | ${nic_name} -| Suite Teardown | Tear down 3-node performance topology -| ... -| Test Setup | Set up performance test -| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd -| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs} -| ... | dut2_node=${dut2} | dut2_vm_refs=${dut2_vm_refs} -| ... -| Test Template | Local Template -| ... -| Documentation | *RFC2544: Pkt throughput L2BD test cases with vhost* -| ... -| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology -| ... | with single links between nodes. -| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 switching of IPv4. -| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 bridge- -| ... | domains and MAC learning enabled. Qemu Guests are connected to VPP via -| ... | vhost-user interfaces. Guests are running DPDK testpmd interconnecting -| ... | vhost-user interfaces using 5 cores pinned to cpus 6-10 and 11-15 and -| ... | 2048M memory. Testpmd is using socket-mem=1024M (512x2M hugepages), -| ... | 5 cores (1 main core and 4 cores dedicated for io), forwarding mode is -| ... | set to io, rxd/txd=1024, burst=64. DUT1, DUT2 are tested -| ... | with ${nic_name}.\ -| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop\ -| ... | Rate) with zero packet loss tolerance and throughput PDR (Partial Drop\ -| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage\ -| ... | of packets transmitted. NDR and PDR are discovered for different\ -| ... | Ethernet L2 frame sizes using MLRsearch library.\ -| ... | Test packets are generated by TG on links to DUTs. TG traffic profile -| ... | contains two L3 flow-groups (flow-group per direction, 254 flows per -| ... | flow-group) with all packets containing Ethernet header, IPv4 header -| ... | with IP protocol=61 and static payload. MAC addresses are matching MAC -| ... | addresses of the TG node interfaces. - -*** Variables *** -| ${nic_name}= | Intel-X710 -| ${overhead}= | ${0} -# Traffic profile: -| ${traffic_profile}= | trex-sl-3n-ethip4-ip4src254 - -*** Keywords *** -| Local Template -| | [Documentation] -| | ... | [Cfg] DUT runs L2BD switching config. -| | ... | Each DUT uses ${phy_cores} physical core(s) for worker threads. -| | ... | [Ver] Measure NDR and PDR values using MLRsearch algorithm.\ -| | ... -| | ... | *Arguments:* -| | ... | - frame_size - Framesize in Bytes in integer or string (IMIX_v4_1). -| | ... | Type: integer, string -| | ... | - phy_cores - Number of physical cores. Type: integer -| | ... | - rxq - Number of RX queues, default value: ${None}. Type: integer -| | ... -| | [Arguments] | ${frame_size} | ${phy_cores} | ${rxq}=${None} -| | ... -| | Set Test Variable | \${frame_size} -| | ... -| | ${dut1_vm_refs}= | Create Dictionary -| | ${dut2_vm_refs}= | Create Dictionary -| | Set Test Variable | ${dut1_vm_refs} -| | Set Test Variable | ${dut2_vm_refs} -| | Given Add worker threads and rxqueues to all DUTs | ${phy_cores} | ${rxq} -| | And Add PCI devices to all DUTs -| | Set Max Rate And Jumbo And Handle Multi Seg -| | And Apply startup configuration on all VPP DUTs -| | When Initialize L2 bridge domains with Vhost-User | nf_nodes=${2} -| | And Configure guest VMs with dpdk-testpmd connected via vhost-user -| | ... | vm_count=${2} | jumbo=${jumbo} | perf_qemu_qsz=${1024} -| | ... | use_tuned_cfs=${False} -| | Then Find NDR and PDR intervals using optimized search - -*** Test Cases *** -| tc01-64B-1c-eth-l2bdbasemaclrn-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 64B | 1C -| | frame_size=${64} | phy_cores=${1} - -| tc02-64B-2c-eth-l2bdbasemaclrn-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 64B | 2C -| | frame_size=${64} | phy_cores=${2} - -| tc03-64B-4c-eth-l2bdbasemaclrn-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 64B | 4C -| | frame_size=${64} | phy_cores=${4} - -| tc04-1518B-1c-eth-l2bdbasemaclrn-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 1518B | 1C -| | frame_size=${1518} | phy_cores=${1} - -| tc05-1518B-2c-eth-l2bdbasemaclrn-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 1518B | 2C -| | frame_size=${1518} | phy_cores=${2} - -| tc06-1518B-4c-eth-l2bdbasemaclrn-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 1518B | 4C -| | frame_size=${1518} | phy_cores=${4} - -| tc07-9000B-1c-eth-l2bdbasemaclrn-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 9000B | 1C -| | frame_size=${9000} | phy_cores=${1} - -| tc08-9000B-2c-eth-l2bdbasemaclrn-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 9000B | 2C -| | frame_size=${9000} | phy_cores=${2} - -| tc09-9000B-4c-eth-l2bdbasemaclrn-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 9000B | 4C -| | frame_size=${9000} | phy_cores=${4} - -| tc10-IMIX-1c-eth-l2bdbasemaclrn-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | IMIX | 1C -| | frame_size=IMIX_v4_1 | phy_cores=${1} - -| tc11-IMIX-2c-eth-l2bdbasemaclrn-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | IMIX | 2C -| | frame_size=IMIX_v4_1 | phy_cores=${2} - -| tc12-IMIX-4c-eth-l2bdbasemaclrn-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | IMIX | 4C -| | frame_size=IMIX_v4_1 | phy_cores=${4} diff --git a/tests/vpp/perf/vm_vhost/10ge2p1x710-eth-l2xcbase-eth-2vhostvr1024-1vm-ndrpdr.robot b/tests/vpp/perf/vm_vhost/10ge2p1x710-eth-l2xcbase-eth-2vhostvr1024-1vm-ndrpdr.robot index 065e323380..1a01032153 100644 --- a/tests/vpp/perf/vm_vhost/10ge2p1x710-eth-l2xcbase-eth-2vhostvr1024-1vm-ndrpdr.robot +++ b/tests/vpp/perf/vm_vhost/10ge2p1x710-eth-l2xcbase-eth-2vhostvr1024-1vm-ndrpdr.robot @@ -22,9 +22,7 @@ | Suite Teardown | Tear down 3-node performance topology | ... | Test Setup | Set up performance test -| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd -| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs} -| ... | dut2_node=${dut2} | dut2_vm_refs=${dut2_vm_refs} +| Test Teardown | Tear down performance test with vhost | ... | Test Template | Local Template | ... @@ -34,12 +32,10 @@ | ... | with single links between nodes. | ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 switching of IPv4. | ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 cross- -| ... | connect. Qemu Guest is connected to VPP via vhost-user interfaces. -| ... | Guest is running DPDK testpmd interconnecting vhost-user interfaces -| ... | using 5 cores pinned to cpus 5-9 and 2048M memory. Testpmd is using -| ... | socket-mem=1024M (512x2M hugepages), 5 cores (1 main core and 4 cores -| ... | dedicated for io), forwarding mode is set to io, rxd/txd=1024, -| ... | burst=64. DUT1, DUT2 are tested with ${nic_name}.\ +| ... | connect. Qemu VNFs are \ +| ... | connected to VPP via vhost-user interfaces. Guest is running VPP l2xc \ +| ... | interconnecting vhost-user interfaces, rxd/txd=1024. DUT1/DUT2 is \ +| ... | tested with ${nic_name}. | ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop\ | ... | Rate) with zero packet loss tolerance and throughput PDR (Partial Drop\ | ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage\ @@ -50,10 +46,15 @@ | ... | flow-group) with all packets containing Ethernet header, IPv4 header | ... | with IP protocol=61 and static payload. MAC addresses are matching MAC | ... | addresses of the TG node interfaces. +| ... | *[Ref] Applicable standard specifications:* RFC2544. *** Variables *** | ${nic_name}= | Intel-X710 | ${overhead}= | ${0} +| ${nf_dtcr}= | ${1} +| ${nf_dtc}= | ${1} +| ${nf_chains}= | ${1} +| ${nf_nodes}= | ${1} # Traffic profile: | ${traffic_profile}= | trex-sl-3n-ethip4-ip4src254 @@ -74,18 +75,14 @@ | | ... | | Set Test Variable | \${frame_size} | | ... -| | ${dut1_vm_refs}= | Create Dictionary -| | ${dut2_vm_refs}= | Create Dictionary -| | Set Test Variable | ${dut1_vm_refs} -| | Set Test Variable | ${dut2_vm_refs} | | Given Add worker threads and rxqueues to all DUTs | ${phy_cores} | ${rxq} | | And Add PCI devices to all DUTs | | Set Max Rate And Jumbo And Handle Multi Seg | | And Apply startup configuration on all VPP DUTs -| | When Initialize L2 xconnect with Vhost-User | vm_count=${1} -| | And Configure guest VMs with dpdk-testpmd connected via vhost-user -| | ... | vm_count=${1} | jumbo=${jumbo} | perf_qemu_qsz=${1024} -| | ... | use_tuned_cfs=${False} +| | When Initialize L2 xconnect with Vhost-User | nf_nodes=${nf_nodes} +| | And Configure chains of NFs connected via vhost-user +| | ... | nf_chains=${nf_chains} | nf_nodes=${nf_nodes} | jumbo=${jumbo} +| | ... | use_tuned_cfs=${False} | auto_scale=${True} | vnf=vpp_chain_l2xc | | Then Find NDR and PDR intervals using optimized search *** Test Cases *** diff --git a/tests/vpp/perf/vm_vhost/10ge2p1x710-eth-l2xcbase-eth-4vhostvr1024-2vm-ndrpdr.robot b/tests/vpp/perf/vm_vhost/10ge2p1x710-eth-l2xcbase-eth-4vhostvr1024-2vm-ndrpdr.robot deleted file mode 100644 index 29249e34f4..0000000000 --- a/tests/vpp/perf/vm_vhost/10ge2p1x710-eth-l2xcbase-eth-4vhostvr1024-2vm-ndrpdr.robot +++ /dev/null @@ -1,138 +0,0 @@ -# Copyright (c) 2019 Cisco and/or its affiliates. -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at: -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. - -*** Settings *** -| Resource | resources/libraries/robot/performance/performance_setup.robot -| ... -| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDR -| ... | NIC_Intel-X710 | ETH | L2XCFWD | BASE | VHOST | 2VM | VHOST_1024 -| ... -| Suite Setup | Set up 3-node performance topology with DUT's NIC model -| ... | L2 | ${nic_name} -| Suite Teardown | Tear down 3-node performance topology -| ... -| Test Setup | Set up performance test -| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd -| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs} -| ... | dut2_node=${dut2} | dut2_vm_refs=${dut2_vm_refs} -| ... -| Test Template | Local Template -| ... -| Documentation | *RFC2544: Pkt throughput L2XC test cases with vhost* -| ... -| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology -| ... | with single links between nodes. -| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 switching of IPv4. -| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 cross- -| ... | connects. Qemu Guests are connected to VPP via vhost-user interfaces. -| ... | Guests are running DPDK testpmd interconnecting vhost-user interfaces -| ... | using 5 cores pinned to cpus 6-10 and 11-15 and 2048M memory. Testpmd is -| ... | using socket-mem=1024M (512x2M hugepages), 5 cores (1 main core and 4 -| ... | cores dedicated for io), forwarding mode is set to io, rxd/txd=1024, -| ... | burst=64. DUT1, DUT2 are tested with ${nic_name}.\ -| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop\ -| ... | Rate) with zero packet loss tolerance and throughput PDR (Partial Drop\ -| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage\ -| ... | of packets transmitted. NDR and PDR are discovered for different\ -| ... | Ethernet L2 frame sizes using MLRsearch library.\ -| ... | Test packets are generated by TG on links to DUTs. TG traffic profile -| ... | contains two L3 flow-groups (flow-group per direction, 254 flows per -| ... | flow-group) with all packets containing Ethernet header, IPv4 header -| ... | with IP protocol=61 and static payload. MAC addresses are matching MAC -| ... | addresses of the TG node interfaces. - -*** Variables *** -| ${nic_name}= | Intel-X710 -| ${overhead}= | ${0} -# Traffic profile: -| ${traffic_profile}= | trex-sl-3n-ethip4-ip4src254 - -*** Keywords *** -| Local Template -| | [Documentation] -| | ... | [Cfg] DUT runs L2XC switching config. -| | ... | Each DUT uses ${phy_cores} physical core(s) for worker threads. -| | ... | [Ver] Measure NDR and PDR values using MLRsearch algorithm.\ -| | ... -| | ... | *Arguments:* -| | ... | - frame_size - Framesize in Bytes in integer or string (IMIX_v4_1). -| | ... | Type: integer, string -| | ... | - phy_cores - Number of physical cores. Type: integer -| | ... | - rxq - Number of RX queues, default value: ${None}. Type: integer -| | ... -| | [Arguments] | ${frame_size} | ${phy_cores} | ${rxq}=${None} -| | ... -| | Set Test Variable | \${frame_size} -| | ... -| | ${dut1_vm_refs}= | Create Dictionary -| | ${dut2_vm_refs}= | Create Dictionary -| | Set Test Variable | ${dut1_vm_refs} -| | Set Test Variable | ${dut2_vm_refs} -| | Given Add worker threads and rxqueues to all DUTs | ${phy_cores} | ${rxq} -| | And Add PCI devices to all DUTs -| | Set Max Rate And Jumbo And Handle Multi Seg -| | And Apply startup configuration on all VPP DUTs -| | When Initialize L2 xconnect with Vhost-User | vm_count=${2} -| | And Configure guest VMs with dpdk-testpmd connected via vhost-user -| | ... | vm_count=${2} | jumbo=${jumbo} | perf_qemu_qsz=${1024} -| | ... | use_tuned_cfs=${False} -| | Then Find NDR and PDR intervals using optimized search - -*** Test Cases *** -| tc01-64B-1c-eth-l2xcbase-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 64B | 1C -| | frame_size=${64} | phy_cores=${1} - -| tc02-64B-2c-eth-l2xcbase-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 64B | 2C -| | frame_size=${64} | phy_cores=${2} - -| tc03-64B-4c-eth-l2xcbase-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 64B | 4C -| | frame_size=${64} | phy_cores=${4} - -| tc04-1518B-1c-eth-l2xcbase-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 1518B | 1C -| | frame_size=${1518} | phy_cores=${1} - -| tc05-1518B-2c-eth-l2xcbase-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 1518B | 2C -| | frame_size=${1518} | phy_cores=${2} - -| tc06-1518B-4c-eth-l2xcbase-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 1518B | 4C -| | frame_size=${1518} | phy_cores=${4} - -| tc07-9000B-1c-eth-l2xcbase-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 9000B | 1C -| | frame_size=${9000} | phy_cores=${1} - -| tc08-9000B-2c-eth-l2xcbase-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 9000B | 2C -| | frame_size=${9000} | phy_cores=${2} - -| tc09-9000B-4c-eth-l2xcbase-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 9000B | 4C -| | frame_size=${9000} | phy_cores=${4} - -| tc10-IMIX-1c-eth-l2xcbase-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | IMIX | 1C -| | frame_size=IMIX_v4_1 | phy_cores=${1} - -| tc11-IMIX-2c-eth-l2xcbase-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | IMIX | 2C -| | frame_size=IMIX_v4_1 | phy_cores=${2} - -| tc12-IMIX-4c-eth-l2xcbase-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | IMIX | 4C -| | frame_size=IMIX_v4_1 | phy_cores=${4} diff --git a/tests/vpp/perf/vm_vhost/10ge2p1x710-ethip4-ip4base-eth-2vhostvr1024-1vm-ndrpdr.robot b/tests/vpp/perf/vm_vhost/10ge2p1x710-ethip4-ip4base-eth-2vhostvr1024-1vm-ndrpdr.robot index d76b2897de..b3f36830ae 100644 --- a/tests/vpp/perf/vm_vhost/10ge2p1x710-ethip4-ip4base-eth-2vhostvr1024-1vm-ndrpdr.robot +++ b/tests/vpp/perf/vm_vhost/10ge2p1x710-ethip4-ip4base-eth-2vhostvr1024-1vm-ndrpdr.robot @@ -22,9 +22,7 @@ | Suite Teardown | Tear down 3-node performance topology | ... | Test Setup | Set up performance test -| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd -| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs} -| ... | dut2_node=${dut2} | dut2_vm_refs=${dut2_vm_refs} +| Test Teardown | Tear down performance test with vhost | ... | Test Template | Local Template | ... @@ -34,13 +32,10 @@ | ... | with single links between nodes. | ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing. | ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4 -| ... | routing and two static IPv4 /24 route entries. Qemu Guest is connected -| ... | to VPP via vhost-user interfaces. Guest is running DPDK testpmd -| ... | interconnecting vhost-user interfaces using 5 cores pinned to cpus 5-9 -| ... | and 2048M memory. Testpmd is using socket-mem=1024M (512x2M hugepages), -| ... | 5 cores (1 main core and 4 cores dedicated for io), forwarding mode is -| ... | set to mac, rxd/txd=1024, burst=64. DUT1, DUT2 are tested -| ... | with ${nic_name}.\ +| ... | routing and two static IPv4 /24 route entries. Qemu VNFs are \ +| ... | connected to VPP via vhost-user interfaces. Guest is running VPP ip4 \ +| ... | interconnecting vhost-user interfaces, rxd/txd=1024. DUT1/DUT2 is \ +| ... | tested with ${nic_name}. | ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop\ | ... | Rate) with zero packet loss tolerance and throughput PDR (Partial Drop\ | ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage\ @@ -51,10 +46,15 @@ | ... | flow-group) with all packets containing Ethernet header, IPv4 header | ... | with IP protocol=61 and static payload. MAC addresses are matching MAC | ... | addresses of the TG node interfaces. +| ... | *[Ref] Applicable standard specifications:* RFC2544. *** Variables *** | ${nic_name}= | Intel-X710 | ${overhead}= | ${0} +| ${nf_dtcr}= | ${1} +| ${nf_dtc}= | ${1} +| ${nf_chains}= | ${1} +| ${nf_nodes}= | ${1} # Traffic profile: | ${traffic_profile}= | trex-sl-3n-ethip4-ip4src253 @@ -75,19 +75,15 @@ | | ... | | Set Test Variable | \${frame_size} | | ... -| | ${dut1_vm_refs}= | Create Dictionary -| | ${dut2_vm_refs}= | Create Dictionary -| | Set Test Variable | ${dut1_vm_refs} -| | Set Test Variable | ${dut2_vm_refs} | | Given Add worker threads and rxqueues to all DUTs | ${phy_cores} | ${rxq} | | And Add PCI devices to all DUTs | | Set Max Rate And Jumbo And Handle Multi Seg | | And Apply startup configuration on all VPP DUTs | | When Initialize IPv4 forwarding with vhost in 3-node circular topology -| | ... | vm_count=${1} -| | And Configure guest VMs with dpdk-testpmd-mac connected via vhost-user -| | ... | vm_count=${1} | jumbo=${jumbo} | perf_qemu_qsz=${1024} -| | ... | use_tuned_cfs=${False} +| | ... | nf_nodes=${nf_nodes} +| | And Configure chains of NFs connected via vhost-user +| | ... | nf_chains=${nf_chains} | nf_nodes=${nf_nodes} | jumbo=${jumbo} +| | ... | use_tuned_cfs=${False} | auto_scale=${True} | vnf=vpp_chain_ip4_noarp | | Then Find NDR and PDR intervals using optimized search *** Test Cases *** diff --git a/tests/vpp/perf/vm_vhost/10ge2p1x710-ethip4-ip4base-eth-4vhostvr1024-2vm-ndrpdr.robot b/tests/vpp/perf/vm_vhost/10ge2p1x710-ethip4-ip4base-eth-4vhostvr1024-2vm-ndrpdr.robot deleted file mode 100644 index ea56bd15a1..0000000000 --- a/tests/vpp/perf/vm_vhost/10ge2p1x710-ethip4-ip4base-eth-4vhostvr1024-2vm-ndrpdr.robot +++ /dev/null @@ -1,140 +0,0 @@ -# Copyright (c) 2019 Cisco and/or its affiliates. -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at: -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. - -*** Settings *** -| Resource | resources/libraries/robot/performance/performance_setup.robot -| ... -| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDR -| ... | NIC_Intel-X710 | ETH | IP4FWD | BASE | VHOST | 2VM | VHOST_1024 -| ... -| Suite Setup | Set up 3-node performance topology with DUT's NIC model -| ... | L3 | ${nic_name} -| Suite Teardown | Tear down 3-node performance topology -| ... -| Test Setup | Set up performance test -| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd -| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs} -| ... | dut2_node=${dut2} | dut2_vm_refs=${dut2_vm_refs} -| ... -| Test Template | Local Template -| ... -| Documentation | *RFC2544: Pkt throughput IPv4 test cases with vhost* -| ... -| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology -| ... | with single links between nodes. -| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing. -| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4 -| ... | routing and two static IPv4 /24 route entries. Qemu Guests are connected -| ... | to VPP via vhost-user interfaces. Guests are running DPDK testpmd -| ... | interconnectingvhost-user interfaces using 5 cores pinned to cpus 6-10 -| ... | and 11-15 and 2048M memory. Testpmd is using socket-mem=1024M (512x2M -| ... | hugepages), 5 cores (1 main core and 4 cores dedicated for io), -| ... | forwarding mode is set to io, rxd/txd=1024, burst=64. DUT1, DUT2 are -| ... | tested with ${nic_name}.\ -| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop\ -| ... | Rate) with zero packet loss tolerance and throughput PDR (Partial Drop\ -| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage\ -| ... | of packets transmitted. NDR and PDR are discovered for different\ -| ... | Ethernet L2 frame sizes using MLRsearch library.\ -| ... | Test packets are generated by TG on links to DUTs. TG traffic profile -| ... | contains two L3 flow-groups (flow-group per direction, 253 flows per -| ... | flow-group) with all packets containing Ethernet header, IPv4 header -| ... | with IP protocol=61 and static payload. MAC addresses are matching MAC -| ... | addresses of the TG node interfaces. - -*** Variables *** -| ${nic_name}= | Intel-X710 -| ${overhead}= | ${0} -# Traffic profile: -| ${traffic_profile}= | trex-sl-3n-ethip4-ip4src253 - -*** Keywords *** -| Local Template -| | [Documentation] -| | ... | [Cfg] DUT runs IPv4 routing config. -| | ... | Each DUT uses ${phy_cores} physical core(s) for worker threads. -| | ... | [Ver] Measure NDR and PDR values using MLRsearch algorithm.\ -| | ... -| | ... | *Arguments:* -| | ... | - frame_size - Framesize in Bytes in integer or string (IMIX_v4_1). -| | ... | Type: integer, string -| | ... | - phy_cores - Number of physical cores. Type: integer -| | ... | - rxq - Number of RX queues, default value: ${None}. Type: integer -| | ... -| | [Arguments] | ${frame_size} | ${phy_cores} | ${rxq}=${None} -| | ... -| | Set Test Variable | \${frame_size} -| | ... -| | ${dut1_vm_refs}= | Create Dictionary -| | ${dut2_vm_refs}= | Create Dictionary -| | Set Test Variable | ${dut1_vm_refs} -| | Set Test Variable | ${dut2_vm_refs} -| | Given Add worker threads and rxqueues to all DUTs | ${phy_cores} | ${rxq} -| | And Add PCI devices to all DUTs -| | Set Max Rate And Jumbo And Handle Multi Seg -| | And Apply startup configuration on all VPP DUTs -| | When Initialize IPv4 forwarding with vhost in 3-node circular topology -| | ... | vm_count=${2} -| | And Configure guest VMs with dpdk-testpmd-mac connected via vhost-user -| | ... | vm_count=${2} | jumbo=${jumbo} | perf_qemu_qsz=${1024} -| | ... | use_tuned_cfs=${False} -| | Then Find NDR and PDR intervals using optimized search - -*** Test Cases *** -| tc01-64B-1c-ethip4-ip4base-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 64B | 1C -| | frame_size=${64} | phy_cores=${1} - -| tc02-64B-2c-ethip4-ip4base-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 64B | 2C -| | frame_size=${64} | phy_cores=${2} - -| tc03-64B-4c-ethip4-ip4base-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 64B | 4C -| | frame_size=${64} | phy_cores=${4} - -| tc04-1518B-1c-ethip4-ip4base-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 1518B | 1C -| | frame_size=${1518} | phy_cores=${1} - -| tc05-1518B-2c-ethip4-ip4base-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 1518B | 2C -| | frame_size=${1518} | phy_cores=${2} - -| tc06-1518B-4c-ethip4-ip4base-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 1518B | 4C -| | frame_size=${1518} | phy_cores=${4} - -| tc07-9000B-1c-ethip4-ip4base-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 9000B | 1C -| | frame_size=${9000} | phy_cores=${1} - -| tc08-9000B-2c-ethip4-ip4base-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 9000B | 2C -| | frame_size=${9000} | phy_cores=${2} - -| tc09-9000B-4c-ethip4-ip4base-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 9000B | 4C -| | frame_size=${9000} | phy_cores=${4} - -| tc10-IMIX-1c-ethip4-ip4base-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | IMIX | 1C -| | frame_size=IMIX_v4_1 | phy_cores=${1} - -| tc11-IMIX-2c-ethip4-ip4base-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | IMIX | 2C -| | frame_size=IMIX_v4_1 | phy_cores=${2} - -| tc12-IMIX-4c-ethip4-ip4base-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | IMIX | 4C -| | frame_size=IMIX_v4_1 | phy_cores=${4} diff --git a/tests/vpp/perf/vm_vhost/10ge2p1x710-ethip4vxlan-l2bdbasemaclrn-eth-2vhostvr1024-1vm-ndrpdr.robot b/tests/vpp/perf/vm_vhost/10ge2p1x710-ethip4vxlan-l2bdbasemaclrn-eth-2vhostvr1024-1vm-ndrpdr.robot index acff0e6337..501917e273 100644 --- a/tests/vpp/perf/vm_vhost/10ge2p1x710-ethip4vxlan-l2bdbasemaclrn-eth-2vhostvr1024-1vm-ndrpdr.robot +++ b/tests/vpp/perf/vm_vhost/10ge2p1x710-ethip4vxlan-l2bdbasemaclrn-eth-2vhostvr1024-1vm-ndrpdr.robot @@ -23,9 +23,7 @@ | Suite Teardown | Tear down 3-node performance topology | ... | Test Setup | Set up performance test -| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd -| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs} -| ... | dut2_node=${dut2} | dut2_vm_refs=${dut2_vm_refs} +| Test Teardown | Tear down performance test with vhost | ... | Test Template | Local Template | ... @@ -37,12 +35,10 @@ | ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 switching of IPv4. | ... | Eth-IPv4-VXLAN-Eth-IPv4 is applied on link between DUT1 and DUT2. | ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 bridge- -| ... | domain and MAC learning enabled. Qemu Guest is connected to VPP via -| ... | vhost-user interfaces. Guest is running DPDK testpmd interconnecting -| ... | vhost-user interfaces using 5 cores pinned to cpus 5-9 and 2048M -| ... | memory. Testpmd is using socket-mem=1024M (512x2M hugepages), 5 cores -| ... | (1 main core and 4 cores dedicated for io), forwarding mode is set to -| ... | io, rxd/txd=1024, burst=64. DUT1, DUT2 are tested with ${nic_name}.\ +| ... | domain and MAC learning enabled. Qemu VNFs are \ +| ... | connected to VPP via vhost-user interfaces. Guest is running VPP l2xc \ +| ... | interconnecting vhost-user interfaces, rxd/txd=1024. DUT1/DUT2 is \ +| ... | tested with ${nic_name}. | ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop\ | ... | Rate) with zero packet loss tolerance and throughput PDR (Partial Drop\ | ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage\ @@ -60,6 +56,10 @@ | ${overhead}= | ${50} | ${bd_id1}= | 1 | ${bd_id2}= | 2 +| ${nf_dtcr}= | ${1} +| ${nf_dtc}= | ${1} +| ${nf_chains}= | ${1} +| ${nf_nodes}= | ${1} # Traffic profile: | ${traffic_profile}= | trex-sl-3n-ethip4-ip4src254 @@ -80,19 +80,15 @@ | | ... | | Set Test Variable | \${frame_size} | | ... -| | ${dut1_vm_refs}= | Create Dictionary -| | ${dut2_vm_refs}= | Create Dictionary -| | Set Test Variable | ${dut1_vm_refs} -| | Set Test Variable | ${dut2_vm_refs} | | Given Add worker threads and rxqueues to all DUTs | ${phy_cores} | ${rxq} | | And Add PCI devices to all DUTs | | Set Max Rate And Jumbo And Handle Multi Seg | | And Apply startup configuration on all VPP DUTs | | When Initialize L2 bridge domains with Vhost-User and VXLANoIPv4 in 3-node circular topology | | ... | ${bd_id1} | ${bd_id2} -| | And Configure guest VMs with dpdk-testpmd connected via vhost-user -| | ... | vm_count=${1} | jumbo=${jumbo} | perf_qemu_qsz=${1024} -| | ... | use_tuned_cfs=${False} +| | And Configure chains of NFs connected via vhost-user +| | ... | nf_chains=${nf_chains} | nf_nodes=${nf_chains} | jumbo=${jumbo} +| | ... | use_tuned_cfs=${False} | auto_scale=${True} | vnf=vpp_chain_l2xc | | Then Find NDR and PDR intervals using optimized search *** Test Cases *** diff --git a/tests/vpp/perf/vm_vhost/2n1l-10ge2p1x710-dot1q-l2bdbasemaclrn-eth-2vhostvr1024-1vm-ndrpdr.robot b/tests/vpp/perf/vm_vhost/2n1l-10ge2p1x710-dot1q-l2bdbasemaclrn-eth-2vhostvr1024-1vm-ndrpdr.robot index e24d9f017e..43d14cdfde 100644 --- a/tests/vpp/perf/vm_vhost/2n1l-10ge2p1x710-dot1q-l2bdbasemaclrn-eth-2vhostvr1024-1vm-ndrpdr.robot +++ b/tests/vpp/perf/vm_vhost/2n1l-10ge2p1x710-dot1q-l2bdbasemaclrn-eth-2vhostvr1024-1vm-ndrpdr.robot @@ -23,8 +23,7 @@ | Suite Teardown | Tear down 2-node performance topology | ... | Test Setup | Set up performance test -| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd -| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs} +| Test Teardown | Tear down performance test with vhost | ... | Test Template | Local Template | ... @@ -36,21 +35,20 @@ | ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 switching of IPv4. IEEE\ | ... | 802.1Q tagging is applied on link between DUT1-if2 and TG-if2. | ... | *[Cfg] DUT configuration:* DUT1 is configured with L2 bridge-domain and\ -| ... | MAC learning enabled. Qemu Guest is connected to VPP via vhost-user\ -| ... | interfaces. Guest is running DPDK testpmd interconnecting vhost-user\ -| ... | interfaces, forwarding mode is set to io, rxd/txd=1024. DUT1 is tested\ -| ... | with ${nic_name}.\ +| ... | MAC learning enabled. Qemu VNFs are connected to VPP via\ +| ... | vhost-user interfaces. Guest is running VPP l2xc interconnecting \ +| ... | vhost-user interfaces, rxd/txd=1024. DUT1 is tested with ${nic_name}. | ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop\ | ... | Rate) with zero packet loss tolerance and throughput PDR (Partial Drop\ | ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage\ | ... | of packets transmitted. NDR and PDR are discovered for different\ | ... | Ethernet L2 frame sizes using MLRsearch library.\ -| ... | Test packets are\ -| ... | generated by TG on links to DUTs. TG traffic profile contains two L3\ -| ... | flow-groups (flow-group per direction, 254 flows per flow-group) with\ -| ... | all packets containing Ethernet header, IPv4 header with IP protocol=61\ -| ... | and static payload. MAC addresses are matching MAC addresses of the TG\ -| ... | node interfaces. +| ... | Test packets are generated by TG on\ +| ... | links to DUTs. TG traffic profile contains two L3 flow-groups\ +| ... | (flow-group per direction, 254 flows per flow-group) with all packets\ +| ... | containing Ethernet header, IPv4 header with IP protocol=61 and static\ +| ... | payload. MAC addresses are matching MAC addresses of NFs nodes\ +| ... | interfaces. | ... | *[Ref] Applicable standard specifications:* RFC2544. *** Variables *** @@ -60,6 +58,10 @@ | ${tag_rewrite}= | pop-1 | ${bd_id1}= | 1 | ${bd_id2}= | 2 +| ${nf_dtcr}= | ${1} +| ${nf_dtc}= | ${1} +| ${nf_chains}= | ${1} +| ${nf_nodes}= | ${1} # Traffic profile: | ${traffic_profile}= | trex-sl-2n-dot1qip4asym-ip4src254 @@ -80,16 +82,15 @@ | | ... | | Set Test Variable | \${frame_size} | | ... -| | ${dut1_vm_refs}= | Create Dictionary -| | Set Test Variable | ${dut1_vm_refs} | | Given Add worker threads and rxqueues to all DUTs | ${phy_cores} | ${rxq} | | And Add PCI devices to all DUTs | | Set Max Rate And Jumbo And Handle Multi Seg | | And Apply startup configuration on all VPP DUTs | | When Initialize L2 bridge domains with Vhost-User and VLAN in circular topology | | ... | ${bd_id1} | ${bd_id2} | ${subid} | ${tag_rewrite} -| | And Configure guest VMs with dpdk-testpmd connected via vhost-user -| | ... | jumbo=${jumbo} | perf_qemu_qsz=${1024} | use_tuned_cfs=${False} +| | And Configure chains of NFs connected via vhost-user +| | ... | nf_chains=${nf_chains} | nf_nodes=${nf_nodes} | jumbo=${jumbo} +| | ... | use_tuned_cfs=${False} | auto_scale=${True} | vnf=vpp_chain_l2xc | | Then Find NDR and PDR intervals using optimized search *** Test Cases *** diff --git a/tests/vpp/perf/vm_vhost/2n1l-10ge2p1x710-eth-l2bdbasemaclrn-eth-2vhostvr1024-1vm-ndrpdr.robot b/tests/vpp/perf/vm_vhost/2n1l-10ge2p1x710-eth-l2bdbasemaclrn-eth-2vhostvr1024-1vm-ndrpdr.robot index 3c4bbe0041..aaecbb665a 100644 --- a/tests/vpp/perf/vm_vhost/2n1l-10ge2p1x710-eth-l2bdbasemaclrn-eth-2vhostvr1024-1vm-ndrpdr.robot +++ b/tests/vpp/perf/vm_vhost/2n1l-10ge2p1x710-eth-l2bdbasemaclrn-eth-2vhostvr1024-1vm-ndrpdr.robot @@ -22,8 +22,7 @@ | Suite Teardown | Tear down 2-node performance topology | ... | Test Setup | Set up performance test -| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd -| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs} +| Test Teardown | Tear down performance test with vhost | ... | Test Template | Local Template | ... @@ -33,10 +32,10 @@ | ... | with single links between nodes. | ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 switching of IPv4. | ... | *[Cfg] DUT configuration:* DUT1 is configured with L2 bridge- -| ... | domain and MAC learning enabled. Qemu Guest is connected to VPP via -| ... | vhost-user interfaces. Guest is running DPDK testpmd -| ... | interconnecting vhost-user interfaces, forwarding mode is -| ... | set to io, rxd/txd=256. DUT1 is tested with ${nic_name}.\ +| ... | domain and MAC learning enabled. Qemu VNFs are connected \ +| ... | to VPP via vhost-user interfaces. Guest is running VPP l2xc \ +| ... | interconnecting vhost-user interfaces, rxd/txd=1024. DUT1 is tested \ +| ... | with ${nic_name}. | ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop\ | ... | Rate) with zero packet loss tolerance and throughput PDR (Partial Drop\ | ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage\ @@ -53,6 +52,10 @@ *** Variables *** | ${nic_name}= | Intel-X710 | ${overhead}= | ${0} +| ${nf_dtcr}= | ${1} +| ${nf_dtc}= | ${1} +| ${nf_chains}= | ${1} +| ${nf_nodes}= | ${1} # Traffic profile: | ${traffic_profile}= | trex-sl-2n-ethip4-ip4src254 @@ -73,16 +76,14 @@ | | ... | | Set Test Variable | \${frame_size} | | ... -| | ${dut1_vm_refs}= | Create Dictionary -| | Set Test Variable | ${dut1_vm_refs} | | Given Add worker threads and rxqueues to all DUTs | ${phy_cores} | ${rxq} | | And Add PCI devices to all DUTs | | Set Max Rate And Jumbo And Handle Multi Seg | | And Apply startup configuration on all VPP DUTs -| | When Initialize L2 bridge domains with Vhost-User | nf_nodes=${1} -| | And Configure guest VMs with dpdk-testpmd connected via vhost-user -| | ... | vm_count=${1} | jumbo=${jumbo} | perf_qemu_qsz=${1024} -| | ... | use_tuned_cfs=${False} +| | When Initialize L2 bridge domains with Vhost-User | nf_nodes=${nf_nodes} +| | And Configure chains of NFs connected via vhost-user +| | ... | nf_chains=${nf_chains} | nf_nodes=${nf_nodes} | jumbo=${jumbo} +| | ... | use_tuned_cfs=${False} | auto_scale=${True} | vnf=vpp_chain_l2xc | | Then Find NDR and PDR intervals using optimized search *** Test Cases *** diff --git a/tests/vpp/perf/vm_vhost/2n1l-10ge2p1x710-eth-l2bdbasemaclrn-eth-4vhostvr1024-2vm-ndrpdr.robot b/tests/vpp/perf/vm_vhost/2n1l-10ge2p1x710-eth-l2bdbasemaclrn-eth-4vhostvr1024-2vm-ndrpdr.robot deleted file mode 100644 index 365bcbacb5..0000000000 --- a/tests/vpp/perf/vm_vhost/2n1l-10ge2p1x710-eth-l2bdbasemaclrn-eth-4vhostvr1024-2vm-ndrpdr.robot +++ /dev/null @@ -1,133 +0,0 @@ -# Copyright (c) 2019 Cisco and/or its affiliates. -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at: -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. - -*** Settings *** -| Resource | resources/libraries/robot/performance/performance_setup.robot -| ... -| Force Tags | 2_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDR -| ... | NIC_Intel-X710 | ETH | L2BDMACLRN | BASE | VHOST | 2VM | VHOST_1024 -| ... -| Suite Setup | Set up 2-node performance topology with DUT's NIC model -| ... | L2 | ${nic_name} -| Suite Teardown | Tear down 2-node performance topology -| ... -| Test Setup | Set up performance test -| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd -| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs} -| ... -| Test Template | Local Template -| ... -| Documentation | *RFC2544: Pkt throughput L2BD test cases with vhost* -| ... -| ... | *[Top] Network Topologies:* TG-DUT1-TG 2-node circular topology -| ... | with single links between nodes. -| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 switching of IPv4. -| ... | *[Cfg] DUT configuration:* DUT1 is configured with L2 bridge- -| ... | domain and MAC learning enabled. Qemu Guest is connected to VPP via -| ... | vhost-user interfaces. Guest is running DPDK testpmd -| ... | interconnecting vhost-user interfaces, forwarding mode is -| ... | set to io, rxd/txd=256. DUT1 is tested with ${nic_name}.\ -| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop\ -| ... | Rate) with zero packet loss tolerance and throughput PDR (Partial Drop\ -| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage\ -| ... | of packets transmitted. NDR and PDR are discovered for different\ -| ... | Ethernet L2 frame sizes using MLRsearch library.\ -| ... | Test packets are generated by TG on links to DUTs. TG traffic profile -| ... | contains two L3 flow-groups (flow-group per direction, 254 flows per -| ... | flow-group) with all packets containing Ethernet header, IPv4 header -| ... | with IP protocol=61 and static payload. MAC addresses are matching MAC -| ... | addresses of the TG node interfaces. - -*** Variables *** -| ${nic_name}= | Intel-X710 -| ${overhead}= | ${0} -# Traffic profile: -| ${traffic_profile}= | trex-sl-2n-ethip4-ip4src254 - -*** Keywords *** -| Local Template -| | [Documentation] -| | ... | [Cfg] DUT runs L2BD switching config. -| | ... | Each DUT uses ${phy_cores} physical core(s) for worker threads. -| | ... | [Ver] Measure NDR and PDR values using MLRsearch algorithm.\ -| | ... -| | ... | *Arguments:* -| | ... | - frame_size - Framesize in Bytes in integer or string (IMIX_v4_1). -| | ... | Type: integer, string -| | ... | - phy_cores - Number of physical cores. Type: integer -| | ... | - rxq - Number of RX queues, default value: ${None}. Type: integer -| | ... -| | [Arguments] | ${frame_size} | ${phy_cores} | ${rxq}=${None} -| | ... -| | Set Test Variable | \${frame_size} -| | ... -| | ${dut1_vm_refs}= | Create Dictionary -| | Set Test Variable | ${dut1_vm_refs} -| | Given Add worker threads and rxqueues to all DUTs | ${phy_cores} | ${rxq} -| | And Add PCI devices to all DUTs -| | Set Max Rate And Jumbo And Handle Multi Seg -| | And Apply startup configuration on all VPP DUTs -| | When Initialize L2 bridge domains with Vhost-User | nf_nodes=${2} -| | And Configure guest VMs with dpdk-testpmd connected via vhost-user -| | ... | vm_count=${2} | jumbo=${jumbo} | perf_qemu_qsz=${1024} -| | ... | use_tuned_cfs=${False} -| | Then Find NDR and PDR intervals using optimized search - -*** Test Cases *** -| tc01-64B-1c-eth-l2bdbasemaclrn-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 64B | 1C -| | frame_size=${64} | phy_cores=${1} - -| tc02-64B-2c-eth-l2bdbasemaclrn-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 64B | 2C -| | frame_size=${64} | phy_cores=${2} - -| tc03-64B-4c-eth-l2bdbasemaclrn-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 64B | 4C -| | frame_size=${64} | phy_cores=${4} - -| tc04-1518B-1c-eth-l2bdbasemaclrn-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 1518B | 1C -| | frame_size=${1518} | phy_cores=${1} - -| tc05-1518B-2c-eth-l2bdbasemaclrn-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 1518B | 2C -| | frame_size=${1518} | phy_cores=${2} - -| tc06-1518B-4c-eth-l2bdbasemaclrn-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 1518B | 4C -| | frame_size=${1518} | phy_cores=${4} - -| tc07-9000B-1c-eth-l2bdbasemaclrn-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 9000B | 1C -| | frame_size=${9000} | phy_cores=${1} - -| tc08-9000B-2c-eth-l2bdbasemaclrn-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 9000B | 2C -| | frame_size=${9000} | phy_cores=${2} - -| tc09-9000B-4c-eth-l2bdbasemaclrn-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 9000B | 4C -| | frame_size=${9000} | phy_cores=${4} - -| tc10-IMIX-1c-eth-l2bdbasemaclrn-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | IMIX | 1C -| | frame_size=IMIX_v4_1 | phy_cores=${1} - -| tc11-IMIX-2c-eth-l2bdbasemaclrn-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | IMIX | 2C -| | frame_size=IMIX_v4_1 | phy_cores=${2} - -| tc12-IMIX-4c-eth-l2bdbasemaclrn-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | IMIX | 4C -| | frame_size=IMIX_v4_1 | phy_cores=${4} diff --git a/tests/vpp/perf/vm_vhost/2n1l-10ge2p1x710-eth-l2xcbase-eth-2vhostvr1024-1vm-ndrpdr.robot b/tests/vpp/perf/vm_vhost/2n1l-10ge2p1x710-eth-l2xcbase-eth-2vhostvr1024-1vm-ndrpdr.robot index af0632e136..e855acad8c 100644 --- a/tests/vpp/perf/vm_vhost/2n1l-10ge2p1x710-eth-l2xcbase-eth-2vhostvr1024-1vm-ndrpdr.robot +++ b/tests/vpp/perf/vm_vhost/2n1l-10ge2p1x710-eth-l2xcbase-eth-2vhostvr1024-1vm-ndrpdr.robot @@ -22,8 +22,7 @@ | Suite Teardown | Tear down 2-node performance topology | ... | Test Setup | Set up performance test -| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd -| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs} +| Test Teardown | Tear down performance test with vhost | ... | Test Template | Local Template | ... @@ -33,10 +32,10 @@ | ... | with single links between nodes. | ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 switching of IPv4. | ... | *[Cfg] DUT configuration:* DUT1 is configured with L2 cross- -| ... | connect. Qemu Guest is connected to VPP via vhost-user interfaces. -| ... | Guest is running DPDK testpmd interconnecting vhost-user interfaces, -| ... | forwarding mode is set to io, rxd/txd=256. DUT1 is tested -| ... | with ${nic_name}.\ +| ... | connect. Qemu VNFs are connected \ +| ... | to VPP via vhost-user interfaces. Guest is running VPP l2xc \ +| ... | interconnecting vhost-user interfaces, rxd/txd=1024. DUT1 is tested \ +| ... | with ${nic_name}. | ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop\ | ... | Rate) with zero packet loss tolerance and throughput PDR (Partial Drop\ | ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage\ @@ -47,10 +46,15 @@ | ... | flow-group) with all packets containing Ethernet header, IPv4 header | ... | with IP protocol=61 and static payload. MAC addresses are matching MAC | ... | addresses of the TG node interfaces. +| ... | *[Ref] Applicable standard specifications:* RFC2544. *** Variables *** | ${nic_name}= | Intel-X710 | ${overhead}= | ${0} +| ${nf_dtcr}= | ${1} +| ${nf_dtc}= | ${1} +| ${nf_chains}= | ${1} +| ${nf_nodes}= | ${1} # Traffic profile: | ${traffic_profile}= | trex-sl-2n-ethip4-ip4src254 @@ -71,16 +75,14 @@ | | ... | | Set Test Variable | \${frame_size} | | ... -| | ${dut1_vm_refs}= | Create Dictionary -| | Set Test Variable | ${dut1_vm_refs} | | Given Add worker threads and rxqueues to all DUTs | ${phy_cores} | ${rxq} | | And Add PCI devices to all DUTs | | Set Max Rate And Jumbo And Handle Multi Seg | | And Apply startup configuration on all VPP DUTs -| | When Initialize L2 xconnect with Vhost-User | vm_count=${1} -| | And Configure guest VMs with dpdk-testpmd connected via vhost-user -| | ... | vm_count=${1} | jumbo=${jumbo} | perf_qemu_qsz=${1024} -| | ... | use_tuned_cfs=${False} +| | When Initialize L2 xconnect with Vhost-User | nf_nodes=${nf_nodes} +| | And Configure chains of NFs connected via vhost-user +| | ... | nf_chains=${nf_chains} | nf_nodes=${nf_nodes} | jumbo=${jumbo} +| | ... | use_tuned_cfs=${False} | auto_scale=${True} | vnf=vpp_chain_l2xc | | Then Find NDR and PDR intervals using optimized search *** Test Cases *** diff --git a/tests/vpp/perf/vm_vhost/2n1l-10ge2p1x710-eth-l2xcbase-eth-4vhostvr1024-2vm-ndrpdr.robot b/tests/vpp/perf/vm_vhost/2n1l-10ge2p1x710-eth-l2xcbase-eth-4vhostvr1024-2vm-ndrpdr.robot deleted file mode 100644 index 195b5d9c9b..0000000000 --- a/tests/vpp/perf/vm_vhost/2n1l-10ge2p1x710-eth-l2xcbase-eth-4vhostvr1024-2vm-ndrpdr.robot +++ /dev/null @@ -1,133 +0,0 @@ -# Copyright (c) 2019 Cisco and/or its affiliates. -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at: -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. - -*** Settings *** -| Resource | resources/libraries/robot/performance/performance_setup.robot -| ... -| Force Tags | 2_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDR -| ... | NIC_Intel-X710 | ETH | L2XCFWD | BASE | VHOST | 2VM | VHOST_1024 -| ... -| Suite Setup | Set up 2-node performance topology with DUT's NIC model -| ... | L2 | ${nic_name} -| Suite Teardown | Tear down 2-node performance topology -| ... -| Test Setup | Set up performance test -| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd -| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs} -| ... -| Test Template | Local Template -| ... -| Documentation | *RFC2544: Pkt throughput L2XC test cases with vhost* -| ... -| ... | *[Top] Network Topologies:* TG-DUT1-TG 2-node circular topology -| ... | with single links between nodes. -| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 switching of IPv4. -| ... | *[Cfg] DUT configuration:* DUT1 is configured with L2 cross- -| ... | connect. Qemu Guest is connected to VPP via vhost-user interfaces. -| ... | Guest is running DPDK testpmd interconnecting vhost-user interfaces, -| ... | forwarding mode is set to io, rxd/txd=256. DUT1 is tested -| ... | with ${nic_name}.\ -| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop\ -| ... | Rate) with zero packet loss tolerance and throughput PDR (Partial Drop\ -| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage\ -| ... | of packets transmitted. NDR and PDR are discovered for different\ -| ... | Ethernet L2 frame sizes using MLRsearch library.\ -| ... | Test packets are generated by TG on links to DUTs. TG traffic profile -| ... | contains two L3 flow-groups (flow-group per direction, 254 flows per -| ... | flow-group) with all packets containing Ethernet header, IPv4 header -| ... | with IP protocol=61 and static payload. MAC addresses are matching MAC -| ... | addresses of the TG node interfaces. - -*** Variables *** -| ${nic_name}= | Intel-X710 -| ${overhead}= | ${0} -# Traffic profile: -| ${traffic_profile}= | trex-sl-2n-ethip4-ip4src254 - -*** Keywords *** -| Local Template -| | [Documentation] -| | ... | [Cfg] DUT runs L2XC switching config. -| | ... | Each DUT uses ${phy_cores} physical core(s) for worker threads. -| | ... | [Ver] Measure NDR and PDR values using MLRsearch algorithm.\ -| | ... -| | ... | *Arguments:* -| | ... | - frame_size - Framesize in Bytes in integer or string (IMIX_v4_1). -| | ... | Type: integer, string -| | ... | - phy_cores - Number of physical cores. Type: integer -| | ... | - rxq - Number of RX queues, default value: ${None}. Type: integer -| | ... -| | [Arguments] | ${frame_size} | ${phy_cores} | ${rxq}=${None} -| | ... -| | Set Test Variable | \${frame_size} -| | ... -| | ${dut1_vm_refs}= | Create Dictionary -| | Set Test Variable | ${dut1_vm_refs} -| | Given Add worker threads and rxqueues to all DUTs | ${phy_cores} | ${rxq} -| | And Add PCI devices to all DUTs -| | Set Max Rate And Jumbo And Handle Multi Seg -| | And Apply startup configuration on all VPP DUTs -| | When Initialize L2 xconnect with Vhost-User | vm_count=${2} -| | And Configure guest VMs with dpdk-testpmd connected via vhost-user -| | ... | vm_count=${2} | jumbo=${jumbo} | perf_qemu_qsz=${1024} -| | ... | use_tuned_cfs=${False} -| | Then Find NDR and PDR intervals using optimized search - -*** Test Cases *** -| tc01-64B-1c-eth-l2xcbase-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 64B | 1C -| | frame_size=${64} | phy_cores=${1} - -| tc02-64B-2c-eth-l2xcbase-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 64B | 2C -| | frame_size=${64} | phy_cores=${2} - -| tc03-64B-4c-eth-l2xcbase-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 64B | 4C -| | frame_size=${64} | phy_cores=${4} - -| tc04-1518B-1c-eth-l2xcbase-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 1518B | 1C -| | frame_size=${1518} | phy_cores=${1} - -| tc05-1518B-2c-eth-l2xcbase-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 1518B | 2C -| | frame_size=${1518} | phy_cores=${2} - -| tc06-1518B-4c-eth-l2xcbase-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 1518B | 4C -| | frame_size=${1518} | phy_cores=${4} - -| tc07-9000B-1c-eth-l2xcbase-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 9000B | 1C -| | frame_size=${9000} | phy_cores=${1} - -| tc08-9000B-2c-eth-l2xcbase-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 9000B | 2C -| | frame_size=${9000} | phy_cores=${2} - -| tc09-9000B-4c-eth-l2xcbase-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 9000B | 4C -| | frame_size=${9000} | phy_cores=${4} - -| tc10-IMIX-1c-eth-l2xcbase-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | IMIX | 1C -| | frame_size=IMIX_v4_1 | phy_cores=${1} - -| tc11-IMIX-2c-eth-l2xcbase-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | IMIX | 2C -| | frame_size=IMIX_v4_1 | phy_cores=${2} - -| tc12-IMIX-4c-eth-l2xcbase-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | IMIX | 4C -| | frame_size=IMIX_v4_1 | phy_cores=${4} diff --git a/tests/vpp/perf/vm_vhost/2n1l-10ge2p1x710-ethip4-ip4base-eth-2vhostvr1024-1vm-ndrpdr.robot b/tests/vpp/perf/vm_vhost/2n1l-10ge2p1x710-ethip4-ip4base-eth-2vhostvr1024-1vm-ndrpdr.robot index 08516c8ff9..ab3350278f 100644 --- a/tests/vpp/perf/vm_vhost/2n1l-10ge2p1x710-ethip4-ip4base-eth-2vhostvr1024-1vm-ndrpdr.robot +++ b/tests/vpp/perf/vm_vhost/2n1l-10ge2p1x710-ethip4-ip4base-eth-2vhostvr1024-1vm-ndrpdr.robot @@ -22,8 +22,7 @@ | Suite Teardown | Tear down 2-node performance topology | ... | Test Setup | Set up performance test -| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd -| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs} +| Test Teardown | Tear down performance test with vhost | ... | Test Template | Local Template | ... @@ -33,10 +32,10 @@ | ... | with single links between nodes. | ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing. | ... | *[Cfg] DUT configuration:* DUT1 is configured with IPv4 -| ... | routing and two static IPv4 /24 route entries. Qemu Guest is connected -| ... | to VPP via vhost-user interfaces. Guest is running DPDK testpmd -| ... | interconnecting vhost-user interfaces, forwarding mode is -| ... | set to mac, rxd/txd=1024. DUT1 is tested with ${nic_name}.\ +| ... | routing and two static IPv4 /24 route entries. Qemu VNFs are connected \ +| ... | to VPP via vhost-user interfaces. Guest is running VPP ip4 \ +| ... | interconnecting vhost-user interfaces, rxd/txd=1024. DUT1 is tested \ +| ... | with ${nic_name}. | ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop\ | ... | Rate) with zero packet loss tolerance and throughput PDR (Partial Drop\ | ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage\ @@ -47,10 +46,15 @@ | ... | flow-group) with all packets containing Ethernet header, IPv4 header | ... | with IP protocol=61 and static payload. MAC addresses are matching MAC | ... | addresses of the TG node interfaces. +| ... | *[Ref] Applicable standard specifications:* RFC2544. *** Variables *** | ${nic_name}= | Intel-X710 | ${overhead}= | ${0} +| ${nf_dtcr}= | ${1} +| ${nf_dtc}= | ${1} +| ${nf_chains}= | ${1} +| ${nf_nodes}= | ${1} # Traffic profile: | ${traffic_profile}= | trex-sl-2n-ethip4-ip4src253 @@ -71,17 +75,15 @@ | | ... | | Set Test Variable | \${frame_size} | | ... -| | ${dut1_vm_refs}= | Create Dictionary -| | Set Test Variable | ${dut1_vm_refs} | | Given Add worker threads and rxqueues to all DUTs | ${phy_cores} | ${rxq} | | And Add PCI devices to all DUTs | | Set Max Rate And Jumbo And Handle Multi Seg | | And Apply startup configuration on all VPP DUTs | | When Initialize IPv4 forwarding with vhost in 2-node circular topology -| | ... | vm_count=${1} -| | And Configure guest VMs with dpdk-testpmd-mac connected via vhost-user -| | ... | vm_count=${1} | jumbo=${jumbo} | perf_qemu_qsz=${1024} -| | ... | use_tuned_cfs=${False} +| | ... | nf_nodes=${nf_nodes} +| | And Configure chains of NFs connected via vhost-user +| | ... | nf_chains=${nf_chains} | nf_nodes=${nf_nodes} | jumbo=${jumbo} +| | ... | use_tuned_cfs=${False} | auto_scale=${True} | vnf=vpp_chain_ip4_noarp | | Then Find NDR and PDR intervals using optimized search *** Test Cases *** diff --git a/tests/vpp/perf/vm_vhost/2n1l-10ge2p1x710-ethip4-ip4base-eth-4vhostvr1024-2vm-ndrpdr.robot b/tests/vpp/perf/vm_vhost/2n1l-10ge2p1x710-ethip4-ip4base-eth-4vhostvr1024-2vm-ndrpdr.robot deleted file mode 100644 index 24a040d219..0000000000 --- a/tests/vpp/perf/vm_vhost/2n1l-10ge2p1x710-ethip4-ip4base-eth-4vhostvr1024-2vm-ndrpdr.robot +++ /dev/null @@ -1,134 +0,0 @@ -# Copyright (c) 2019 Cisco and/or its affiliates. -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at: -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. - -*** Settings *** -| Resource | resources/libraries/robot/performance/performance_setup.robot -| ... -| Force Tags | 2_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDR -| ... | NIC_Intel-X710 | ETH | IP4FWD | BASE | VHOST | 2VM | VHOST_1024 -| ... -| Suite Setup | Set up 2-node performance topology with DUT's NIC model -| ... | L3 | ${nic_name} -| Suite Teardown | Tear down 2-node performance topology -| ... -| Test Setup | Set up performance test -| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd -| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs} -| ... -| Test Template | Local Template -| ... -| Documentation | *RFC2544: Pkt throughput IPv4 test cases with vhost* -| ... -| ... | *[Top] Network Topologies:* TG-DUT1-TG 2-node circular topology -| ... | with single links between nodes. -| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing. -| ... | *[Cfg] DUT configuration:* DUT1 is configured with IPv4 -| ... | routing and two static IPv4 /24 route entries. Qemu Guest is connected -| ... | to VPP via vhost-user interfaces. Guest is running DPDK testpmd -| ... | interconnecting vhost-user interfaces, forwarding mode is -| ... | set to mac, rxd/txd=1024. DUT1 is tested with ${nic_name}.\ -| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop\ -| ... | Rate) with zero packet loss tolerance and throughput PDR (Partial Drop\ -| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage\ -| ... | of packets transmitted. NDR and PDR are discovered for different\ -| ... | Ethernet L2 frame sizes using MLRsearch library.\ -| ... | Test packets are generated by TG on links to DUTs. TG traffic profile -| ... | contains two L3 flow-groups (flow-group per direction, 253 flows per -| ... | flow-group) with all packets containing Ethernet header, IPv4 header -| ... | with IP protocol=61 and static payload. MAC addresses are matching MAC -| ... | addresses of the TG node interfaces. - -*** Variables *** -| ${nic_name}= | Intel-X710 -| ${overhead}= | ${0} -# Traffic profile: -| ${traffic_profile}= | trex-sl-2n-ethip4-ip4src253 - -*** Keywords *** -| Local Template -| | [Documentation] -| | ... | [Cfg] DUT runs IPv4 routing config. -| | ... | Each DUT uses ${phy_cores} physical core(s) for worker threads. -| | ... | [Ver] Measure NDR and PDR values using MLRsearch algorithm.\ -| | ... -| | ... | *Arguments:* -| | ... | - frame_size - Framesize in Bytes in integer or string (IMIX_v4_1). -| | ... | Type: integer, string -| | ... | - phy_cores - Number of physical cores. Type: integer -| | ... | - rxq - Number of RX queues, default value: ${None}. Type: integer -| | ... -| | [Arguments] | ${frame_size} | ${phy_cores} | ${rxq}=${None} -| | ... -| | Set Test Variable | \${frame_size} -| | ... -| | ${dut1_vm_refs}= | Create Dictionary -| | Set Test Variable | ${dut1_vm_refs} -| | Given Add worker threads and rxqueues to all DUTs | ${phy_cores} | ${rxq} -| | And Add PCI devices to all DUTs -| | Set Max Rate And Jumbo And Handle Multi Seg -| | And Apply startup configuration on all VPP DUTs -| | When Initialize IPv4 forwarding with vhost in 2-node circular topology -| | ... | vm_count=${2} -| | And Configure guest VMs with dpdk-testpmd-mac connected via vhost-user -| | ... | vm_count=${2} | jumbo=${jumbo} | perf_qemu_qsz=${1024} -| | ... | use_tuned_cfs=${False} -| | Then Find NDR and PDR intervals using optimized search - -*** Test Cases *** -| tc01-64B-1c-ethip4-ip4base-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 64B | 1C -| | frame_size=${64} | phy_cores=${1} - -| tc02-64B-2c-ethip4-ip4base-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 64B | 2C -| | frame_size=${64} | phy_cores=${2} - -| tc03-64B-4c-ethip4-ip4base-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 64B | 4C -| | frame_size=${64} | phy_cores=${4} - -| tc04-1518B-1c-ethip4-ip4base-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 1518B | 1C -| | frame_size=${1518} | phy_cores=${1} - -| tc05-1518B-2c-ethip4-ip4base-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 1518B | 2C -| | frame_size=${1518} | phy_cores=${2} - -| tc06-1518B-4c-ethip4-ip4base-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 1518B | 4C -| | frame_size=${1518} | phy_cores=${4} - -| tc07-9000B-1c-ethip4-ip4base-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 9000B | 1C -| | frame_size=${9000} | phy_cores=${1} - -| tc08-9000B-2c-ethip4-ip4base-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 9000B | 2C -| | frame_size=${9000} | phy_cores=${2} - -| tc09-9000B-4c-ethip4-ip4base-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | 9000B | 4C -| | frame_size=${9000} | phy_cores=${4} - -| tc10-IMIX-1c-ethip4-ip4base-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | IMIX | 1C -| | frame_size=IMIX_v4_1 | phy_cores=${1} - -| tc11-IMIX-2c-ethip4-ip4base-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | IMIX | 2C -| | frame_size=IMIX_v4_1 | phy_cores=${2} - -| tc12-IMIX-4c-ethip4-ip4base-eth-4vhostvr1024-2vm-ndrpdr -| | [Tags] | IMIX | 4C -| | frame_size=IMIX_v4_1 | phy_cores=${4} diff --git a/tests/vpp/perf/vts/10ge2p1x710-ethip4vxlan-l2bdbasemaclrn-eth-iacldstbase-aclpermit-2vhostvr1024-1vm-ndrpdr.robot b/tests/vpp/perf/vts/10ge2p1x710-ethip4vxlan-l2bdbasemaclrn-eth-iacldstbase-aclpermit-2vhostvr1024-1vm-ndrpdr.robot index b5119e08fc..27ef3203dc 100644 --- a/tests/vpp/perf/vts/10ge2p1x710-ethip4vxlan-l2bdbasemaclrn-eth-iacldstbase-aclpermit-2vhostvr1024-1vm-ndrpdr.robot +++ b/tests/vpp/perf/vts/10ge2p1x710-ethip4vxlan-l2bdbasemaclrn-eth-iacldstbase-aclpermit-2vhostvr1024-1vm-ndrpdr.robot @@ -25,8 +25,7 @@ | Suite Teardown | Tear down 3-node performance topology | ... | Test Setup | Set up performance test -| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd and ACL -| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs} +| Test Teardown | Tear down performance test with vhost | ... | Test Template | Local Template | ... @@ -38,12 +37,10 @@ | ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 switching of IPv4. | ... | Eth-IPv4-VXLAN-Eth-IPv4 is applied on link between DUT1 and DUT2. | ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 bridge- -| ... | domain and MAC learning enabled. Qemu Guest is connected to VPP via -| ... | vhost-user interfaces. Guest is running DPDK testpmd interconnecting -| ... | vhost-user interfaces using 5 cores pinned to cpus 5-9 and 2048M -| ... | memory. Testpmd is using socket-mem=1024M (512x2M hugepages), 5 cores -| ... | (1 main core and 4 cores dedicated for io), forwarding mode is set to -| ... | io, rxd/txd=256, burst=64. DUT1, DUT2 are tested with ${nic_name}.\ +| ... | domain and MAC learning enabled. Qemu VNFs are connected \ +| ... | to VPP via vhost-user interfaces. Guest is running VPP l2xc \ +| ... | interconnecting vhost-user interfaces, rxd/txd=1024. DUT1/DUT2 is \ +| ... | tested with ${nic_name}. | ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop\ | ... | Rate) with zero packet loss tolerance and throughput PDR (Partial Drop\ | ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage\ @@ -60,18 +57,16 @@ *** Variables *** | ${nic_name}= | Intel-X710 | ${overhead}= | ${50} -# Socket names +| ${nf_dtcr}= | ${1} +| ${nf_dtc}= | ${1} +| ${nf_chains}= | ${1} +| ${nf_nodes}= | ${1} | ${dut1_bd_id1}= | 1 | ${dut1_bd_id2}= | 2 | ${dut2_bd_id1}= | 1 -| ${sock1}= | /var/run/vpp/sock-1-${dut1_bd_id1} -| ${sock2}= | /var/run/vpp/sock-1-${dut1_bd_id2} # Traffic profile: | ${traffic_profile}= | trex-sl-ethip4-vxlansrc253 | ${acl_type}= | permit -# Defaults for teardown: -| ${dut1}= | ${None} -| ${dut1_vm_refs}= | ${None} *** Keywords *** | Local Template @@ -99,8 +94,6 @@ | | @{dut1_vxlans} = | Create List | ${vxlan1} | | @{dut2_vxlans} = | Create List | ${vxlan2} | | Set interfaces in path up -| | Configure vhost interfaces for L2BD forwarding | ${dut1} -| | ... | ${sock1} | ${sock2} | | When Init L2 bridge domains with single DUT with Vhost-User and VXLANoIPv4 in 3-node circular topology | | ... | 172.16.0.1 | 16 | 172.26.0.1 | 16 | 172.16.0.2 | 172.26.0.2 | | ... | ${dut1_vxlans} | ${dut2_vxlans} | 172.17.0.0 | 16 | 172.27.0.0 | 16 @@ -108,13 +101,10 @@ | | Run Keyword If | '${acl_type}' != '${EMPTY}' | | ... | Configure ACLs on a single interface | ${dut1} | ${dut1_if2} | input | | ... | ${acl_type} | @{permit_list} -| | ${nf_cpus}= | Get Affinity NF | ${nodes} | DUT1 | nf_chains=${1} -| | | ... | nf_nodes=${1} | nf_chain=${1} | nf_node=${1} -| | | ... | vs_dtc=${cpu_count_int} | nf_dtc=${cpu_count_int} -| | ${vm1} = | And Configure guest VM with dpdk-testpmd connected via vhost-user -| | ... | DUT1 | ${sock1} | ${sock2} | ${TEST NAME}DUT1_VM1 | ${nf_cpus} -| | ... | jumbo=${jumbo} | perf_qemu_qsz=${1024} | use_tuned_cfs=${False} -| | Set Test Variable | &{dut1_vm_refs} | ${TEST NAME}DUT1_VM1=${vm1} +| | And Configure chains of NFs connected via vhost-user on single node +| | ... | node=DUT1 | nf_chains=${nf_chains} | nf_nodes=${nf_nodes} +| | ... | jumbo=${jumbo} | use_tuned_cfs=${False} | auto_scale=${True} +| | ... | vnf=vpp_chain_l2xc | | Then Find NDR and PDR intervals using optimized search *** Test Cases *** diff --git a/tests/vpp/perf/vts/10ge2p1x710-ethip4vxlan-l2bdbasemaclrn-eth-iacldstbase-aclpermitreflect-2vhostvr1024-1vm-ndrpdr.robot b/tests/vpp/perf/vts/10ge2p1x710-ethip4vxlan-l2bdbasemaclrn-eth-iacldstbase-aclpermitreflect-2vhostvr1024-1vm-ndrpdr.robot index 0345ceec3c..8e30354eb3 100644 --- a/tests/vpp/perf/vts/10ge2p1x710-ethip4vxlan-l2bdbasemaclrn-eth-iacldstbase-aclpermitreflect-2vhostvr1024-1vm-ndrpdr.robot +++ b/tests/vpp/perf/vts/10ge2p1x710-ethip4vxlan-l2bdbasemaclrn-eth-iacldstbase-aclpermitreflect-2vhostvr1024-1vm-ndrpdr.robot @@ -25,8 +25,7 @@ | Suite Teardown | Tear down 3-node performance topology | ... | Test Setup | Set up performance test -| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd and ACL -| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs} +| Test Teardown | Tear down performance test with vhost | ... | Test Template | Local Template | ... @@ -38,12 +37,10 @@ | ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 switching of IPv4. | ... | Eth-IPv4-VXLAN-Eth-IPv4 is applied on link between DUT1 and DUT2. | ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 bridge- -| ... | domain and MAC learning enabled. Qemu Guest is connected to VPP via -| ... | vhost-user interfaces. Guest is running DPDK testpmd interconnecting -| ... | vhost-user interfaces using 5 cores pinned to cpus 5-9 and 2048M -| ... | memory. Testpmd is using socket-mem=1024M (512x2M hugepages), 5 cores -| ... | (1 main core and 4 cores dedicated for io), forwarding mode is set to -| ... | io, rxd/txd=256, burst=64. DUT1, DUT2 are tested with ${nic_name}.\ +| ... | domain and MAC learning enabled. Qemu VNFs are connected \ +| ... | to VPP via vhost-user interfaces. Guest is running VPP l2xc \ +| ... | interconnecting vhost-user interfaces, rxd/txd=1024. DUT1/DUT2 is \ +| ... | tested with ${nic_name}. | ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop\ | ... | Rate) with zero packet loss tolerance and throughput PDR (Partial Drop\ | ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage\ @@ -60,18 +57,16 @@ *** Variables *** | ${nic_name}= | Intel-X710 | ${overhead}= | ${50} -# Socket names +| ${nf_dtcr}= | ${1} +| ${nf_dtc}= | ${1} +| ${nf_chains}= | ${1} +| ${nf_nodes}= | ${1} | ${dut1_bd_id1}= | 1 | ${dut1_bd_id2}= | 2 | ${dut2_bd_id1}= | 1 -| ${sock1}= | /var/run/vpp/sock-1-${dut1_bd_id1} -| ${sock2}= | /var/run/vpp/sock-1-${dut1_bd_id2} # Traffic profile: | ${traffic_profile}= | trex-sl-ethip4-vxlansrc253 | ${acl_type}= | permit+reflect -# Defaults for teardown: -| ${dut1}= | ${None} -| ${dut1_vm_refs}= | ${None} *** Keywords *** | Local Template @@ -99,22 +94,14 @@ | | @{dut1_vxlans} = | Create List | ${vxlan1} | | @{dut2_vxlans} = | Create List | ${vxlan2} | | Set interfaces in path up -| | Configure vhost interfaces for L2BD forwarding | ${dut1} -| | ... | ${sock1} | ${sock2} | | When Init L2 bridge domains with single DUT with Vhost-User and VXLANoIPv4 in 3-node circular topology | | ... | 172.16.0.1 | 16 | 172.26.0.1 | 16 | 172.16.0.2 | 172.26.0.2 | | ... | ${dut1_vxlans} | ${dut2_vxlans} | 172.17.0.0 | 16 | 172.27.0.0 | 16 | | @{permit_list} = | Create List | 10.0.0.1/32 | 10.0.0.2/32 -| | Run Keyword If | '${acl_type}' != '${EMPTY}' -| | ... | Configure ACLs on a single interface | ${dut1} | ${dut1_if2} | input -| | ... | ${acl_type} | @{permit_list} -| | ${nf_cpus}= | Get Affinity NF | ${nodes} | DUT1 | nf_chains=${1} -| | | ... | nf_nodes=${1} | nf_chain=${1} | nf_node=${1} -| | | ... | vs_dtc=${cpu_count_int} | nf_dtc=${cpu_count_int} -| | ${vm1} = | And Configure guest VM with dpdk-testpmd connected via vhost-user -| | ... | DUT1 | ${sock1} | ${sock2} | ${TEST NAME}DUT1_VM1 | ${nf_cpus} -| | ... | jumbo=${jumbo} | perf_qemu_qsz=${1024} | use_tuned_cfs=${False} -| | Set Test Variable | &{dut1_vm_refs} | ${TEST NAME}DUT1_VM1=${vm1} +| | And Configure chains of NFs connected via vhost-user on single node +| | ... | node=DUT1 | nf_chains=${nf_chains} | nf_nodes=${nf_nodes} +| | ... | jumbo=${jumbo} | use_tuned_cfs=${False} | auto_scale=${True} +| | ... | vnf=vpp_chain_l2xc | | Then Find NDR and PDR intervals using optimized search *** Test Cases *** diff --git a/tests/vpp/perf/vts/10ge2p1x710-ethip4vxlan-l2bdbasemaclrn-eth-iacldstbase-noacl-2vhostvr1024-1vm-ndrpdr.robot b/tests/vpp/perf/vts/10ge2p1x710-ethip4vxlan-l2bdbasemaclrn-eth-iacldstbase-noacl-2vhostvr1024-1vm-ndrpdr.robot index 3b7bd8f921..4e5d4df008 100644 --- a/tests/vpp/perf/vts/10ge2p1x710-ethip4vxlan-l2bdbasemaclrn-eth-iacldstbase-noacl-2vhostvr1024-1vm-ndrpdr.robot +++ b/tests/vpp/perf/vts/10ge2p1x710-ethip4vxlan-l2bdbasemaclrn-eth-iacldstbase-noacl-2vhostvr1024-1vm-ndrpdr.robot @@ -25,8 +25,7 @@ | Suite Teardown | Tear down 3-node performance topology | ... | Test Setup | Set up performance test -| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd and ACL -| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs} +| Test Teardown | Tear down performance test with vhost | ... | Test Template | Local Template | ... @@ -38,12 +37,10 @@ | ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 switching of IPv4. | ... | Eth-IPv4-VXLAN-Eth-IPv4 is applied on link between DUT1 and DUT2. | ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 bridge- -| ... | domain and MAC learning enabled. Qemu Guest is connected to VPP via -| ... | vhost-user interfaces. Guest is running DPDK testpmd interconnecting -| ... | vhost-user interfaces using 5 cores pinned to cpus 5-9 and 2048M -| ... | memory. Testpmd is using socket-mem=1024M (512x2M hugepages), 5 cores -| ... | (1 main core and 4 cores dedicated for io), forwarding mode is set to -| ... | io, rxd/txd=256, burst=64. DUT1, DUT2 are tested with ${nic_name}.\ +| ... | domain and MAC learning enabled. Qemu VNFs are connected \ +| ... | to VPP via vhost-user interfaces. Guest is running VPP l2xc \ +| ... | interconnecting vhost-user interfaces, rxd/txd=1024. DUT1/DUT2 is \ +| ... | tested with ${nic_name}. | ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop\ | ... | Rate) with zero packet loss tolerance and throughput PDR (Partial Drop\ | ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage\ @@ -60,18 +57,16 @@ *** Variables *** | ${nic_name}= | Intel-X710 | ${overhead}= | ${50} -# Socket names +| ${nf_dtcr}= | ${1} +| ${nf_dtc}= | ${1} +| ${nf_chains}= | ${1} +| ${nf_nodes}= | ${1} | ${dut1_bd_id1}= | 1 | ${dut1_bd_id2}= | 2 | ${dut2_bd_id1}= | 1 -| ${sock1}= | /var/run/vpp/sock-1-${dut1_bd_id1} -| ${sock2}= | /var/run/vpp/sock-1-${dut1_bd_id2} # Traffic profile: | ${traffic_profile}= | trex-sl-ethip4-vxlansrc253 | ${acl_type}= | ${EMPTY} -# Defaults for teardown: -| ${dut1}= | ${None} -| ${dut1_vm_refs}= | ${None} *** Keywords *** | Local Template @@ -99,8 +94,6 @@ | | @{dut1_vxlans} = | Create List | ${vxlan1} | | @{dut2_vxlans} = | Create List | ${vxlan2} | | Set interfaces in path up -| | Configure vhost interfaces for L2BD forwarding | ${dut1} -| | ... | ${sock1} | ${sock2} | | When Init L2 bridge domains with single DUT with Vhost-User and VXLANoIPv4 in 3-node circular topology | | ... | 172.16.0.1 | 16 | 172.26.0.1 | 16 | 172.16.0.2 | 172.26.0.2 | | ... | ${dut1_vxlans} | ${dut2_vxlans} | 172.17.0.0 | 16 | 172.27.0.0 | 16 @@ -108,13 +101,10 @@ | | Run Keyword If | '${acl_type}' != '${EMPTY}' | | ... | Configure ACLs on a single interface | ${dut1} | ${dut1_if2} | input | | ... | ${acl_type} | @{permit_list} -| | ${nf_cpus}= | Get Affinity NF | ${nodes} | DUT1 | nf_chains=${1} -| | | ... | nf_nodes=${1} | nf_chain=${1} | nf_node=${1} -| | | ... | vs_dtc=${cpu_count_int} | nf_dtc=${cpu_count_int} -| | ${vm1} = | And Configure guest VM with dpdk-testpmd connected via vhost-user -| | ... | DUT1 | ${sock1} | ${sock2} | ${TEST NAME}DUT1_VM1 | ${nf_cpus} -| | ... | jumbo=${jumbo} | perf_qemu_qsz=${1024} | use_tuned_cfs=${False} -| | Set Test Variable | &{dut1_vm_refs} | ${TEST NAME}DUT1_VM1=${vm1} +| | And Configure chains of NFs connected via vhost-user on single node +| | ... | node=DUT1 | nf_chains=${nf_chains} | nf_nodes=${nf_nodes} +| | ... | jumbo=${jumbo} | use_tuned_cfs=${False} | auto_scale=${True} +| | ... | vnf=vpp_chain_l2xc | | Then Find NDR and PDR intervals using optimized search *** Test Cases *** |